<%BANNER%>

Virtues Unfulfilled

Permanent Link: http://ufdc.ufl.edu/UFE0043818/00001

Material Information

Title: Virtues Unfulfilled the Effects of Land Value Taxation in Three Pennsylvanian Cities
Physical Description: 1 online resource (154 p.)
Language: english
Creator: Murphy, Robert J, Jr
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: altoona -- housing -- income -- land -- pittsburgh -- policy -- property -- tax -- taxation -- value -- wilkes-barre
Urban and Regional Planning -- Dissertations, Academic -- UF
Genre: Urban and Regional Planning thesis, M.A.U.R.P.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In most municipalities across the country, property tax is a prominent direct source of public finance. Generally speaking, it combines a few attributes of property; land, improvements to land and, in some cases personal property. Detractors of the traditional property tax scheme argue that it can encourage land speculation, induce vacant lots in urban cores, and can discourage property owners from improving their properties. A reform to such an entrenched institution is land value taxation, where land is taxed at a higher rate than improvements. Land speculation is thus discouraged as property owners face a tax regardless of how they improve their property. Landowners are encouraged to seek to create the most utility out of their property or sell the property to someone who will because their taxes do not increase based on improvements. By ascertaining the viability of such a tax structure, economic development planners and politicians will be better able to decide if implementation of land value taxation policy is a worthwhile endeavor. In this research, before-and-after analysis was conducted utilizing 1990 and 2000 U.S. Decennial Census data as well as 2005-2009 American Community Survey (ACS) to determine how the theoretical effects of the land value taxation on income and housing characteristics manifest themselves in three Pennsylvanian cities. These cities are Altoona, Pennsylvania, a city that adopted a land value taxation policy in 2002; Pittsburgh, Pennsylvania, a city that adopted land value taxation in 1913 but rescinded it in 2001; and Wilkes-Barre, Pennsylvania, a city that has never adopted land value taxation. Theoretical expectations of the effects of land value taxation were directly compared to the observations in each study area. It was expected that the results of the study would show that the theoretical effects of the policy were reflected in observed data from the case study areas. However, a categorical examination of the ten variables revealed that the theoretical expectations matched observations in only one instance. Therefore, it cannot be determined that in this examination the theoretical effects of land value taxation manifested themselves in U.S. Census and ACS data. As a result, the virtues extolled about the policy were left unfulfilled.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Robert J Murphy.
Thesis: Thesis (M.A.U.R.P.)--University of Florida, 2011.
Local: Adviser: Blanco, Andre.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043818:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043818/00001

Material Information

Title: Virtues Unfulfilled the Effects of Land Value Taxation in Three Pennsylvanian Cities
Physical Description: 1 online resource (154 p.)
Language: english
Creator: Murphy, Robert J, Jr
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: altoona -- housing -- income -- land -- pittsburgh -- policy -- property -- tax -- taxation -- value -- wilkes-barre
Urban and Regional Planning -- Dissertations, Academic -- UF
Genre: Urban and Regional Planning thesis, M.A.U.R.P.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In most municipalities across the country, property tax is a prominent direct source of public finance. Generally speaking, it combines a few attributes of property; land, improvements to land and, in some cases personal property. Detractors of the traditional property tax scheme argue that it can encourage land speculation, induce vacant lots in urban cores, and can discourage property owners from improving their properties. A reform to such an entrenched institution is land value taxation, where land is taxed at a higher rate than improvements. Land speculation is thus discouraged as property owners face a tax regardless of how they improve their property. Landowners are encouraged to seek to create the most utility out of their property or sell the property to someone who will because their taxes do not increase based on improvements. By ascertaining the viability of such a tax structure, economic development planners and politicians will be better able to decide if implementation of land value taxation policy is a worthwhile endeavor. In this research, before-and-after analysis was conducted utilizing 1990 and 2000 U.S. Decennial Census data as well as 2005-2009 American Community Survey (ACS) to determine how the theoretical effects of the land value taxation on income and housing characteristics manifest themselves in three Pennsylvanian cities. These cities are Altoona, Pennsylvania, a city that adopted a land value taxation policy in 2002; Pittsburgh, Pennsylvania, a city that adopted land value taxation in 1913 but rescinded it in 2001; and Wilkes-Barre, Pennsylvania, a city that has never adopted land value taxation. Theoretical expectations of the effects of land value taxation were directly compared to the observations in each study area. It was expected that the results of the study would show that the theoretical effects of the policy were reflected in observed data from the case study areas. However, a categorical examination of the ten variables revealed that the theoretical expectations matched observations in only one instance. Therefore, it cannot be determined that in this examination the theoretical effects of land value taxation manifested themselves in U.S. Census and ACS data. As a result, the virtues extolled about the policy were left unfulfilled.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Robert J Murphy.
Thesis: Thesis (M.A.U.R.P.)--University of Florida, 2011.
Local: Adviser: Blanco, Andre.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043818:00001


This item has the following downloads:


Full Text

PAGE 1

1 VIRTUES UNFULFILLED : THE EFFECTS OF LAND VALUE TAXAT ION IN THREE PENNSYLVANIAN CITIES By ROBERT J. MURPHY, JR. A THESIS PRESENTED TO THE GRADUATE SCHOOL AT THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN URBAN AND REGIONAL PLANNING UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Robert J. Murphy, Jr.

PAGE 3

3 To Mom, Dad, family, and friends for their support and encouragement

PAGE 4

4 ACKNOWLEDGEMENTS I would like to thank my chair, Dr. A ndres Blanco, for his guidance, feedback and patience without which the accomplishment of this thesis would not have been possible. I would also like to thank my other committee members, Dr. Dawn Jourdan and Dr. David Ling, for carefully prodding aspects of this thesis to help improve its overall quality Gibbons, Eric Hilliker, and my parents Bob and Barb Murph y among others, who have offered their opinions and guid ance on this research when asked.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ .............................. 4 LIST OF TABLES ................................ ................................ ................................ ........... 9 LIST OF FIGURES ................................ ................................ ................................ ...... 10 LIST OF ABBREVIATION S ................................ ................................ .......................... 11 ABSTRACT ................................ ................................ ................................ .................. 12 C H A P T E R 1 INTRODUCTION ................................ ................................ ................................ ... 14 Problem Statement ................................ ................................ ................................ 14 Definition of Land Value Taxation ................................ ................................ .......... 15 Case Study Logic ................................ ................................ ................................ .. 16 Summary ................................ ................................ ................................ ............... 17 2 LITERATURE REVIEW ................................ ................................ ......................... 19 Land Value Taxation: An Alternative to the Traditional Property Tax ..................... 19 Land Taxation ................................ ................................ ................................ ........ 20 The Uniqueness of Taxing Land: An Economic Perspective ................................ .. 20 Land Intensity ................................ ................................ ................................ .. 20 Incidence ................................ ................................ ................................ ......... 21 Neutrality ................................ ................................ ................................ ......... 23 Speculation ................................ ................................ ................................ ..... 24 Land Development Pattern Implications ................................ .......................... 25 Consequences to Landowners ................................ ................................ ........ 27 Summary ................................ ................................ ................................ ......... 29 What is Land Value Taxation? ................................ ................................ ............... 30 International Implementation ................................ ................................ ........... 31 Implementation in the Uni ted States ................................ ................................ 32 New York ................................ ................................ ................................ .. 32 Hawaii ................................ ................................ ................................ ....... 33 Pennsylvania ................................ ................................ ............................ 33 Previous Studies of Land Value Taxation ................................ .............................. 36 Theoretical Models ................................ ................................ .......................... 36 General Equilibrium Models ................................ ................................ ............ 39 Regression Models ................................ ................................ ......................... 41 Comparison Approach ................................ ................................ ..................... 44

PAGE 6

6 Summary of Previous Studies ................................ ................................ ......... 44 Relevance to Planning ................................ ................................ ........................... 45 3 METHODOLOGY ................................ ................................ ................................ .. 48 Study Approach ................................ ................................ ................................ ..... 48 Case Studies ................................ ................................ ................................ ... 49 Study Areas ................................ ................................ ................................ .... 51 Altoona ................................ ................................ ................................ ..... 51 Pittsburgh ................................ ................................ ................................ 52 Wilkes Barre ................................ ................................ ............................. 53 Cross Sectional Study ................................ ................................ ..................... 54 Before And After Study Design ................................ ................................ ....... 54 Data Collection ................................ ................................ ................................ ...... 55 U.S. Census of Population and Housing and American Community Survey .... 55 Bureau of Labor Statistics Data ................................ ................................ ....... 58 Limitations ................................ ................................ ................................ ............. 59 Summary ................................ ................................ ................................ ............... 63 4 RESULTS AND ANALYSIS ................................ ................................ ................... 66 Overview ................................ ................................ ................................ ............... 66 Housing Variables ................................ ................................ ................................ 66 Variable Total Housing Units ................................ ................................ ........ 66 Expectations based on theory ................................ ................................ ... 66 Observed ................................ ................................ ................................ .. 67 Determination ................................ ................................ ........................... 68 Variable Occupancy Status ................................ ................................ .......... 69 Expectations based on theory ................................ ................................ ... 69 Observed ................................ ................................ ................................ .. 69 Determination ................................ ................................ ........................... 69 Variable Units in Structure ................................ ................................ ............ 70 Expectations based on theory ................................ ................................ ... 70 Observed ................................ ................................ ................................ .. 71 Determination ................................ ................................ ........................... 71 Variable Property Value ................................ ................................ ............... 72 Expectations based on theory ................................ ................................ ... 72 Observed ................................ ................................ ................................ .. 73 Determination ................................ ................................ ........................... 73 Variable Gross Rent ................................ ................................ ..................... 74 Expectations based on theory ................................ ................................ ... 74 Observed ................................ ................................ ................................ .. 75 Determination ................................ ................................ ........................... 75 Variable Year Built ................................ ................................ ........................ 76 Expectations based on theory ................................ ................................ ... 76 Observed ................................ ................................ ................................ .. 77 Determination ................................ ................................ ........................... 77

PAGE 7

7 Variable Density Measures ................................ ................................ ........... 78 Expectations based on theory ................................ ................................ ... 78 Observed ................................ ................................ ................................ .. 79 Determination ................................ ................................ ........................... 79 Income Variables ................................ ................................ ................................ ... 80 Variable Household Income ................................ ................................ ......... 81 Expectations based on theory ................................ ................................ ... 81 Observed ................................ ................................ ................................ .. 81 Determination ................................ ................................ ........................... 82 Variable Per Capita Income ................................ ................................ .......... 83 Expectations based on theory ................................ ................................ ... 83 Observed ................................ ................................ ................................ .. 84 Determination ................................ ................................ ........................... 84 Variable Income to Poverty Ratio ................................ ................................ 85 Expectations based on theory ................................ ................................ ... 85 Observed ................................ ................................ ................................ .. 85 Determination ................................ ................................ ........................... 86 Summary ................................ ................................ ................................ ............... 86 5 FINDINGS AND CONCLU SIONS ................................ ................................ ........ 113 Implications for Planning ................................ ................................ ...................... 116 Future Research ................................ ................................ ................................ .. 118 A P P E N D I X : DATA REVIEW & ECONOM IC PROFILES ................................ ............ 120 U.S. Census and American Community Survey Analysis ................................ ..... 120 Housing Figures ................................ ................................ ............................ 121 Total housing units ................................ ................................ .................. 121 Occupancy status ................................ ................................ ................... 121 Units in structure ................................ ................................ ..................... 121 Value of owner occupied housing units ................................ ................... 122 Gross rents ................................ ................................ ............................. 123 Year built ................................ ................................ ................................ 123 Density measures ................................ ................................ ................... 125 Income Figures ................................ ................................ ............................. 125 Household Income ................................ ................................ .................. 126 Per Capita Income ................................ ................................ .................. 126 Ratio of Income to Poverty Level ................................ ............................ 126 Bureau of Labor Statistics Data Analysis ................................ ............................. 127 Altoona, PA MSA: ................................ ................................ ......................... 127 Sectors of future growth in the Altoo na, PA MSA: ................................ ......... 128 Heath care and social assistance: ................................ ........................... 128 Transportation and warehousing: ................................ ............................ 128 Administration, support and waste management services: ...................... 129 Pittsburgh, PA MSA: ................................ ................................ ..................... 129 Sectors of future growth for the Pittsburgh, PA MSA: ................................ .... 130

PAGE 8

8 Construction: ................................ ................................ ........................... 130 Manufacturing: ................................ ................................ ........................ 130 Education: ................................ ................................ ............................... 131 Health care and social assistance: ................................ .......................... 131 Scranton Wilkes Barre, PA MSA: ................................ ................................ .. 131 Sectors of future growth in the Scranton Wilkes Barre, PA MSA: .................. 132 Transportation and warehousing: ................................ ............................ 1 32 Management of companies and enterprises: ................................ .......... 133 Accommodation and food services: ................................ ........................ 133 BLS Analysis Conclusions ................................ ................................ ................... 134 Summary ................................ ................................ ................................ ............. 136 LIST OF REFERENCES ................................ ................................ ............................ 149 BIOGRAPHICAL SKETCH ................................ ................................ ......................... 154

PAGE 9

9 LIST OF TABLES T able P age 3 1 List of variables tested in before and after model ................................ .............. 65 4 1 Total housing units ................................ ................................ ............................ 88 4 2 Occupancy status ................................ ................................ .............................. 89 4 3 Units in struc ture ................................ ................................ ............................... 90 4 4 Value of specified owner occupied housing units ................................ .............. 92 4 5. Gross rent ................................ ................................ ................................ ......... 95 4 6 Year structure built ................................ ................................ ............................ 98 4 8 Household income ................................ ................................ ........................... 102 4 9 Median household income ................................ ................................ ............... 108 4 10 Per capita income ................................ ................................ ............................ 109 4 11 Ratio of in come to poverty level ................................ ................................ ....... 110 A 1 Employment figures by industry ................................ ................................ ....... 138 A 2 Location quotients in study area MSAs ................................ ............................ 142 A 3 Shift share analysis ................................ ................................ ......................... 143 A 4 Esteban Marquillas extension ................................ ................................ ......... 145 A 5 List of variables for an attempted population density regression model ........... 148

PAGE 10

10 LIST OF FIGURES Figure P age 2 1 Tax e ffects on an u pward s loping s upply ................................ ........................... 47 2 2 Tax e ffects on a f ixed s upply ................................ ................................ ............. 47 3 1 Before and a fter s tudy d esign ................................ ................................ ........... 64 3 2 Measurement of c hange through a b efore and a fter d esign .............................. 64

PAGE 11

11 LIST OF ABBREVIATION S ACS American Community Survey BLS U S Department of Labor, Bureau of Labor Statistics MSA Metropolitan Statistical Area NY New York PA Pennsylvania US United States

PAGE 12

12 Abst ract of T h e s i s Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of M a s t e r o f A r t s i n U r b a n a n d R e g i o n a l P l a n n i n g VIRTUES UNFU LFILLED: THE EFFECTS OF LAND VALUE TAXATION IN THREE PENNSYLVA NIA N CITIES By Robert J. Murphy, Jr. December 2011 Chair: Dr. Andres Blanco Major: Urban and Regional Planning In most municipalities across the country, property tax is a prominent direct source of public finance. Generally speaking, it combines a few attributes of property; land, improvements to land and, in some cases personal property. Detractors of the traditional property tax scheme argue that it can encourage land speculation, induce vacant lots in urban cores, and can discourage property owners from improving their properties. A reform to such an entrenched institution is land v alue taxation, where land is taxed at a higher rate than improvements Land speculation is thus discouraged as property owners face a tax regardless of how they improve their property. Landowners are encouraged to seek to create the most utility out of their property or sell t he property to someone who will because their taxes do not increase based on improvements. By ascertaining the viability of such a tax structure, e conomic development planners and politicians will be better able to decide if implementation of land value taxation policy is a worthwhile endeavor.

PAGE 13

13 In this research, b efore and after analysis was conducted utilizing 1990 and 2000 U S Decennial Census dat a as well as 2005 2009 American Community Survey (ACS) to determine how the theoretical effects of the land value taxation on income and housing characteristics manifest themselves in three Pennsylvanian cities. These cities are Altoona, Pennsylvania a c ity that adopted a land value taxation policy in 2002 ; Pittsburgh, Pennsylvania a city that adopted land value taxation in 1913 but rescinded it in 2001 ; and Wilkes Barre, Pennsylvania a city that has never adopted land value taxation Theoretical expect ations of the effects of land value taxation w ere direct ly compared to the observations in each study area. It was expected that the result s of the study would show that the theoretical effects of the policy were reflected in observed data from the case s tudy areas. However, a categorical examination of the ten variables revealed that the theoretical expectations matched observations in only one instance. Therefore, it cannot be determined that in this examination the theoretical effects of land value tax ation manifest ed themselv es in U S Census and ACS data. As a result, the virtues extolled about the policy were left unfulfilled.

PAGE 14

14 CHAPTER 1 INTRODUCTION Problem Statement While people across the United States may have differing opinions on politics, th e best football teams, or the best picture at the Oscars, one thing just about everyone can agree on is their dislike of taxes. No matter the manner in which taxes support government services, they seem to be universally pann ed. Nonetheless while jeer ed by many a taxpayer, property taxes are unique. According to Nobel Laureate Economist William Vickrey worst taxes we have. The portion of the tax that falls on sites or land values is the only ma jor tax that is reasonably free of distortionary effects and is not intolerably regressive. The taxes on improvements and personal property are more difficult to assess properly. They impose excess burdens through undue di scouragement of such investment ( Vickrey, 1999 [ as cited in Dye & England, 2009 ] ) Land value taxation captures the v irtuous side of the dichotomous nature of property tax. Its ability to place the entire burden of the tax squarely on the landowner and its lack of distortionary effect s o n economic choices are two characteristics of the policy that lead to its appeal (Oates & Schwab, 2009) Today, property taxes are an ever present issue as the United States pursues economic recovery in the wake of recession. For example, i n June 2011, th e State of New York, for example, passed legislation in support of a c ap on property tax, traditional property tax in this particular instance as a way to remove a government imposed impediment to economic development and consumer spending. As an alterna tive, land

PAGE 15

15 value taxation may provide the desired economic stimulus without some of the negative side effects caused by traditional property taxation. In order to assess the virtues of land value taxation as an alternative to traditional property tax, they must be compared with real world observations to consider the This thesis will consider how theoretical effects of land value taxation on selected income and housing characteristics manifest themselves in U S Census and American Community Survey figures. Definition of Land Value Taxation In the majority of municipalities nationwide, property tax is a prominent direct source of public finance (England, 2007) Generally speaking, it combin es a few attributes of property: land, impr ovements to land, and in some cases personal property. A tax appraiser determines the value of the real property to be taxed by the municipality. The appraiser does this by using a variety of market and cost based valuation methods, analyzing all thr ee attributes. Despite its prominence, traditional property tax is not without its problems. I n some cases this property tax scheme can encourage land speculation and, in turn, allow for the persistence vacant lots in urban cores. It also can discourage property owners from improving their properties as it will increase the tax levied upon them. As a result, property owners are not always encouraged to generate the most utility out of their land and production moves to where the tax scheme allows for the least costs ceteris paribus This traditional form of property taxation has been in place in much of the county for over a century and, in many cases, has become part of legislative inertia. A reform to such an entrenched institution is land value taxat ion, where land value if not the only element of real property taxed, is taxed more than improvement value

PAGE 16

16 (Dye & England, 2009). Within such a tax system, l and speculation is theoretically discouraged as property owners face a sizable tax regardless of how they improve their property. Therefore, l andowners are encouraged to develop their parcels in a way that generates the mos t utility because their taxes are fixed to land value and do not increase based on improvements they make to the property. The land value tax is also theoretically a more economically efficient tax scheme than property tax because demand for the taxable item is not reduced. For example, demand for constr uction services and materials are reduced when those improvements to land are taxed in a traditional property tax system. As a result, a switch from a traditional property tax to a land value tax should, theoretically, lead to reduced urban blight, increased wealth generation and increased economic efficiency. Case Study Logic Wh ile, t echniques for property tax reform and redistribution have be en implemented in Pennsylvania (PA) nationwide utilization of this redevelopment strateg y has yet to come to fruition (Bourassa, 2009) As a result, this thesis will examine a trio of case study cities in the Keystone State: Altoona, Wilkes Barre, and Pittsburgh. The City of Altoona is loca ted in central Pennsylvania with a popul ation of 46,320 (2010 U S Census) A former railroad hub between Philadelphia and Pittsburgh, Altoona has experi enced overall economic and population d ecline since peak s in the mid 20 th century Recently, resurgence in employment has taken place over the last 20 years. The c ity adopted a land value taxation policy in 2002. The City of Wilkes Barre is located in no rtheast ern Pennsylvania with a population of 41, 498 (2010 U S Census) ndustrially driven city, Wilkes Barre is of similar size and economic base to Altoona with the notable exception of its

PAGE 17

17 proximity to Scranton a 75,000 res ident city Wilkes Barre has never implemented land value taxation. The City of Pit tsburgh is located in western Pennsylvania with a population of 305,704 (2010 U S Census) A former powerhouse in the steel industry until the latter 20 th century, Pittsburgh s hares the industrial heritages of Alt oona and Wilkes Barre albeit at a much larger scale. The Steel City adopted a land value taxation policy in 1913 but reverted to a traditional property tax in 2001. Examining these three case studies in particular will enable capture of externalities associated with state and federal property tax regulation as well as mitigating demograp hic differences. As already mentioned, b efore and after case study analysis of changes in income housing development, and economi c trends will be conducted using 1990, 2000, and 2009 data Concurrently this data will be scrutinized while keeping in mind land value taxation adoption and removal dates in the selected case study municipalities. As a result, this analysis will provid e insights into the imme diate effects of the tax policy over the past decade. Summary Chapter 2 reviews relevant literature, including background information on land value taxation along with the development impacts of property tax policy in general Ch apter 3 describes the methodology used to answer the research questions and discusses the limitations of this approach Chapter 4 presents the analysi s of how land value taxation, or the lack thereof, affected selected income and housing characteristics in the three aforementioned case study municipalities. To do this, the expected theoretical effects of the implementation of land value taxation will be categorically compared w ith act ual observed results. Finally, Chapter 5 offers guidance and

PAGE 18

18 suggestions to planners and planning scholars for future research in understanding the influences of land value taxation in municipalities in Pennsylvania and potentially across the United States.

PAGE 19

19 CHAPTER 2 LITERATURE REVIEW Land Value Taxation: An Alternative to th e Traditional Property Tax In this chapter, relevant literature pertaining to the effects of land value taxation on housing and income characteristics will be discussed. It is important to understand the arguments and theoretical debate surrounding land v alue taxation because it is difficult to conduct experimental scientific research on the subject within a laboratory as Arlo ( Anderson, 2009 ) Additionally to this point, opportunities for new study of real world instances of land value taxation adoption are limited, especially in the United States (U. S.) where the practice is essentially confined to municipalities in Pennsylvania and formerly in Hawaii (Bourassa, 2009) Regrettably for research purposes Pittsbu rgh, the sole major metropolitan area under a land value tax regi me rescinded the system in 2001. Anderson (2009) concludes that t his left a void in turns of opportunities to study large markets currently under land value taxation but on the bright side a lso opened the ability to conduct before and after analysis on the Steel City, an opportunity which this paper will later delve into. In examining this literature, a general description of land value taxation will be covered along wi th debates shedding lig ht upon s e ffects on markets, the incidence of the tax, the timing of developme nt and ultimately on housing and income characteristic s. It will also describe previous studies on the impact of land value taxation on housing and income charac teristic s and the relevance to planning of such studies including this paper.

PAGE 20

20 Land Taxation In order to better grasp land value taxation, it is helpful to be able to compare it with traditional property tax as a baseline. Traditionally, property tax is a form of municip al income generation in which millage rates are attached to raw land value and improvements value. These two rates are traditionally set as equal. Sum of the raw land value and the value of improvements to the property multiplied by the mi llage rate results in the property tax paid by a landowner (Dye & England, 2009) Simply put, the difference between traditional property tax and land value tax is that the latter places a majority, if not all, of the mill age rate on the value of land. T hat said, the following section will cover economic theory pertaining to land taxation in general will be discussed followed by a synopsis of land value taxation economic theory and discussion of land value taxation implementation in the United States (U. S.) and abroad. The Uniqueness of Taxing Land: An Economic Perspective A change from a traditional property tax where land and improvements are taxed at an equal rate to a split rate tax produces some interesting effects. Oates and Schwab eloquently outli ne these effects in Land Value Taxation: Theory, Evidence and Practice (2009). This section will draw considerably An alytics of Land Value Taxation Before examining the 2009 Oates & Schwab section, where tax incidence, ne utrality, speculation, and land development patterns will be discussed, the idea of Land Intensity Central to analysis of land value taxation is the idea of land intensity. Land intensity is calculated as the ratio of land value to total property value (Plummer, 2009)

PAGE 21

21 For instance, using what is known as the residual land value method for example, if a parcel is valued at $100,000 and improvements to the parcel amount to $60,000, the be $40,000, the total property value subtracted by the improvement value. In the case of this parcel, its land intensity would be a ratio of 40,000/100,000, or .4. When land value taxation policies are implemented, or any change in how land value or impro vements are taxed, land intensity plays a pivotal role in determining the taxation, a heavier tax on land value versus improvements, owners with an above average land intensity will experience a tax increase. Conversely, owners with below average land intensity will experience a tax decrease as the brunt of the tax rate falls on land value while they hold most of their value in improvements. Also of note is that land intensity is the ratio land value, not land size, to total property value. While land size may often be correlated with land value, large parcel size does not always translate to high land value as land value is impacted by many factors aside from area s uch as loc ation, use or physical features (Plummer, 2009) Keeping the notion of land intensity will be helpful in grasping several of the other tangible effects of land value taxation. Broadly categorized, these impacts fall under tax incidence, neutrali ty, speculation and land development patterns. Incidence seem straight forward, the question belies its complexity as the burden of a tax may not fall upon whom it is levi ed. For instance, an excise tax on any specified commodity may not be borne entirely by the producer, onto whom the tax is levied. Oftentimes, the tax

PAGE 22

22 may be passed on from producers to consume rs in the form of higher prices (Oates & Schwab, 2009) A tax on land acts differently than a tax on standard commodities. Most commodities have an upward sloping supply curve whereas the supply of total land is fixed in quantity and therefore has a vertical supply curve. The same reasoning does not hold true pert do not have a fixed supply. Figure 2 1 shows the tax effects on an upward sloping supply. In this instance, if a tax equal to the distance between A and C is levied on purchases of a g ood, the demand curve will shift down by the amount AB to Demand 2. The new equilibrium at point B shows that lower quantity supplied resulted from lower demand resulting from the higher price induced by the tax. As a result, the burden of the tax is sha red by the consumer, who sees prices rise to P3 for the good which includes the tax, and by the producer, who sees net price received fall to P2 (Oates & Schwab, 2009) A tax on total land, however, beh aves like Figure 2 2 with a fixed supply. While you c perfectly inelastic. As the graph shows, a tax on land is not passed on to the consum er and is borne entirely by the landown er (Oates & Schwab, 2009)

PAGE 23

23 Neutrality In the larger scheme, researchers and policy makers s hare an underlying concern when considering tax policies. According to Oates & Schwab (2009), t the particul ar tax raise sufficient revenue without distorting economic activities by free market system, how will the tax policy generate revenue to cover gaps left by the free marke t without interfering with the free market in a significant way? For example, as was mentioned earlier, an excise tax drives up the cost of production to a producer. The producer then passes this cost to consumers to help share this burden and mai ntain a higher profit margin. This increased price leads to lower consumption of the commodity leading to less wealth ge neration for society as a whole (Oates & Schwab, 2009) ight occurred because the tax has induced a higher price, lowering consumption below an economically efficient level. As was shown in the discussion on tax incidence, land valu e taxation is also unique in regards to tax neutrality. In Figures 2 1, a tax on a commodity in which supply is not fixed results in a price to consumers of P3 yet yields a return to producers of just P2. The tax has lowered the quantity produced from Q1 to Q2 creating a deadweight loss to the economic system and meaning this tax is non neutral. The impact of a tax on land value is again shown in Figure 2 2 Because supply of land is virtually fixed, the quantity produced does not shift and the burden of the tax cannot be shifted to consumers. Thus the land market is not distorted by an increase in

PAGE 24

24 consumer prices and a reduction in supply. Therefore, land taxation in its purest form is way by not inducing deadweight loss (Oates & Schwab, 2009) Speculation One of the central tenants for support of land value taxation is its ability to curb land speculation, thus encouraging infill development rather than vacant lots in downtowns or other developing areas. In 2004, t he Congress for New Urbanism president and CEO and mayor of Milwaukee from 1988 2003, John Norquist, raved about this function e mpty lot in downtown Pitts burgh (Huhne, 2004) On the other hand it has already been stated that land taxation is a neutral tax and does not distort economic choices. What can explain these contradictory points? Th e o retically, regarding land spec ulation holds true A ssuming that the not actual use of land is being tax ed a conversion to land value taxation policy will not hasten development (Tideman, 1982) Cost benefit analysis shows that both traditional property tax and land value tax subtract the same from the present value of both the choice to develop now and the choice to postpone development. However, one caveat arises in this issue. If the landowner does not have the funds to pay the tax currently, the neutralit y of the tax with regards to timing of development comes into question. Liquidity problems may force landowners to either sell the land or develop a revenue stream from it sooner than otherwise in order to pay the land value tax (Oates & Schwab, 2009) Th is effect was explored by Bourassa (1990) as will be discussed later.

PAGE 25

25 Land Development Pattern Implications With concern always growing over the human footprint on our natural environment, concern about urban sprawl is always in the rise as well. Its sign ificance is also part of the debate over the virtues of land value taxation. Would an additional tax on land value cause rural land owners on the urban fringe to develop their properties? In considering the relationship between land value taxation and spr awl, Oates & Schwab (2009) maintain that theoretically it does not. Returning to the discussion on tax neutrality, assuming the tax is placed according to highest and best use and assuming the liquidity is not an issue for the landowner, land value taxa tion is neutral with regards to timing of development. In this way, land value taxation does not induce sprawl. From a different angle, Brueckner and Kim (2003) studied the effects on urban expansion of the traditional property tax, where tax rates on improvement value and land value are equal They noted two effects that run in opposite directions. First, a higher tax r ate on improvements discourages t he production of improvements and lowers density. Lower densities mean that more land is required for the housing of the fixed population, meaning that the traditional property tax increases sprawl. Second, a higher tax r ate on improvements raises the price of improvements, such as housing. If less housing is consumed as a result of the higher total price of improvements, then less land area is required for housing consumpti on, assuming a fixed population. This would mean that the traditional property tax reduces sprawl. Ultimately, because t hese two effects are the most direct in terms of chan ging the size of the urban area and push in opposite directions a myriad of other factors determine which effect is the stronger i n

PAGE 26

26 a particular area. The breadth of these factors is too cumbersome to be delved into here. Extending the Brueckner and Kim analysis, land value taxation is an extreme version of traditional property tax. By placing increased tax rate burden on land value and reducing it on improvement value, land value taxation result in the maximum force on the consumption of land effect, significantly reducing it, and the minimum force on the price of improvements effect, by significantly reducing it as well. As a resu lt, the two counter acting forces described by Brueckner and Kim (2003) would be dominated by a force reducing the urban land area. Theoretically, land value taxation should reduce sprawl. E. Mills (1998) offers a counter perspective to the Brueckner & Ki m study along the lines of curbing sprawl and encouraging infill development. Examining business improvement, Mills produces a two fold answer concerning land value taxation. He holds that replacing the property tax with a la nd tax would increase structu re to land ratios. Additionally, the land tax would increase employment to land ratios This means that the density of buildings and jobs would increase if the traditional property tax is replaced by a land value tax. However, that increase in employment and business activity per unit area will subsequently cause cities to enlarge and sprawl as more workers move to the city faster than building height increases. According to Mills, land value taxation will cause densification of cities in the short term and the enlargement of cities in the long term. State University of New York at Albany Professor Thomas L. Daniels shares Mills opinion on sprawl but offers a different take in that cities today have changed since

PAGE 27

27 ccess to amenities, transportation, and public investment in schools and sewer and water. Access is value. Land value taxation alone will not limit the sprea d of suburbs, because public in vestment in roads, schools, and sewer and water is causing the un ea rned increments on nearby land to rise. Land value taxation applies to those rising land values, thus creating an incenti ve for the land to be (Daniels, 2001) This conclusion is also reached by Turnbull (1988.) Daniels therefore suggests tha t land value taxation must be combined with urban growth boundaries or transfer of development rights to achieve in fill development. In sum, the conclusions of these studies differ because of the frameworks used in each. Brueckner and Kim (2003) studied the housing market as their framework for understanding the impact of land value taxation on sprawl and concluded that it reduces urban land area. E. Mills (1998) used business improvements as the backdrop for his study and found that land value taxation increases density in the short term but increases urban area in the long term. Daniels (2001) and Turnbull (1988) reached a similar conclusion to Mills (1998) but did so under different frameworks as well. Consequences to Landowners While the previous sect ions covering land intensity, tax incidence, neutrality, speculation, and land development patterns focused on large scale economic concerns, this section will cover both the big picture and impacts on landowners, those who are directly impacted by a chang e in property tax policy and those would likely be voting for or against such a change in tax regime. land value taxation literature (Oates & Schwab, 2009) Testing whether a revenue neutr al tax shift,

PAGE 28

28 in which the tax rate on land was increased, required a reduction in the rate on improvements, Brueckner uncovered some significant findings. Intuitively, one would think that the tax rate on improvements must be reduced in order to keep to tal tax revenue unchanged Brueckner concluded that this it is not necessari ly so. He reasoned that in a scenario in which the tax rate on land value was increased, developers would substitute improvements for land. The potential and likelihood exists th at developers may build so many additional buildings that total taxes collected on improvements may even rise despite a lowered tax rate. As a result, a revenue neutral shift from a traditional property tax to land value taxation does not necessarily requ ire a reduction in the tax rate on improvements. In this reasoning, he found that the sensitivity of the supply of improvements to changes in its tax rate to be a key factor in determining whether a reduction in the tax rate on improvements would be necess ary to maintain revenue neutrality Finally, Brueckner examined this elasticity and concluded that the city would likely be required to lower the tax rate on impr ovements despite his hypothesis (Brueckner, 1986) Brueckner would also address the impact of a switch to a split rate property tax system on individual landowners. Despite experiencing a higher tax rate on land, landowners may find reason to favor land value taxation in the form of increased land rents. Land rents are residual profits left over after labor and capital have been attributed their app ropriate shares of profit (Oates & Schwab, 2009) He reasoned that, as previously discussed, the shift from traditional property tax to land value taxation will theoretically result in a reduction in deadweight loss. In turn, this reduction in deadweight loss should yield a more efficient economic system producing higher land

PAGE 29

29 rents. In other words, what was wealth that was lost due to an inefficient economic system becomes surplus value captured by t he landowner. In sum, Brueckner found that in the case of a single community, the impact of reduced deadweight loss more than makes up for the higher tax on land resulting in increased land values. In the case of an entire market, la nd values would likel y decrease (Brueckner, 1986) More on this Previous Studies of Land Value Taxation Summary In thi s section we have seen that a shift from traditional property tax to a split rate t ax in which a higher rate is place on land value than improvement value produces many effects. First off, the concept of land intensity has great bearing over these effects. Unlike excise taxes on a particular commodity, the burden of a tax on land value falls directly on the producer upon whom it is levied and cannot be passed on to final demand The land value tax is a neutral tax and reduces deadweight loss by avoiding economic distortions of higher prices for consumers and reduced supply. A higher t ax on land value does not, theoretically, distort economic decisions. Therefore, it does not affect speculation and timing of development In addition, land value taxation should not theoretically induce sprawl land development patterns. Lastly, while landowners with low land intensity stand to gain from a switch to land value taxation, landowners of high land intensity parcels do not universally stand to lose. Due to increased land rents as the result of gains in economic efficiency property owners implementation, the increased land rents may make up for the higher tax on land.

PAGE 30

30 Generally speaking, the implementation of land value taxation should result in a mo re efficient economic system. With that brief discussion of the theoretical effects of land value taxation we now turn to a synopsis of its implementation in the United States and globally. What is Land Value Taxation? Now that the anomalies of taxing land have been explored, we now turn our attention to what exactly land va lue taxation is. While confusion about this term often abounds, land value taxation exists in our everyday lives in the vast majority of America. Property tax is typically levied equal ly on land and improvements. As Vickrey iterated, in general land value taxation captures the virtuous element of dichotomous nature Its ability to place the entire burden of the tax squarely on the landowner and its aforementioned lack o f distortionary effect on economic choices are two characteristics o f the policy lead to its appeal (Oates & Schwab, 2009) From an equity standpoint, it captures economic surplus produced from the effects of urbanization, not specifically from the effort s of the landowner. This means that the tax captures the effects of efforts from the community as a whole. Thus, it merely returns the collateral public ly generated benefits back to the public. Plummer (2009) illustrates this argument succinctly. Land v alues increase over time because of population growth and community improvements made by the government or private sector (e.g., utility infrastructure, transportation). Taxing land value generates revenue that can benefit the community that provided the individual landowners with their unearned increases in land value. Historically, theories of land value taxation and its virtues began to be explored over the last 150 years. Henry Ge orge with his work Progress and Poverty (1879) was

PAGE 31

31 responsible for b ringing land value taxation to public attention in the late 19 th century. To George, land markets behaved like oligopolistic cartels pushing labor and capital to lands of low quality. As with any cartel behavior, this operation decreased production, jobs wages, and interest rates but raised profits for land owners this was to tax land value to break the cartel. In turn, this would increase production, create jobs, and raise the wage rate, producti on levels, and living standards (Gaff ney, 2008) Additionally, taxing land value would return speculative surplus back to the public domain where it can be to further the aforementioned positive economic indicators. International Implementation Today, although its usage is not as ubiquitous la nd value taxation policies persist around the world. Forms of land taxation can be found from Great Britain to New Zealand and from Japan to Jamaica. infancy as a nation, generating accurate land only valu es for tax assessment was simpler as most of the territory was undeveloped. In time, howe ver, densely developed territories such as Victoria have moved away from tax schemes based on unimproved value to site value and ultimately toward capital improvement or total value tax policies. Site value taxation is an inter mediary policy between unimproved value and capital the value of clearing, leveling, or dr aining, as merged with the land ( Franzsen, 2009) Former Soviet territory Estonia has employed land value taxation in rebuilding its economy since its independence in the early 1990s. Intended as an interim tax policy, land value taxation was implemented here before both land and capital markets had sufficiently established themselves in the post Communist regime (Franzsen, 2009)

PAGE 32

32 According to Tiits (2008), as referenced in Franzsen (2009), due to the underpriced land values produced by the immature land markets at the policies adoption, revenue from the tax could be substantially higher. Implementation in the United States In the United States, land value taxation has appeared in three states ( New York, Hawaii, and Pennsylvania ) but maintains a significant presence only in Pennsylvania. T the City of Amsterdam, adopted the policy from 1995 1996 after the state passed legislation in 1993. Hawaii adopted land value taxation statewide in 1963. In 1913, legislation was passed i n Pennsylvania to allow the states two second tier cities, Pittsburgh and Scranton, to tax buildings and improvement at a lower rate than land. Subsequently, further legislation was passed by the mid 20 th century to allow other municipalities and taxing d istricts to adopt forms of land value taxation. Among the localities to take advantage of this legislation were Harrisburg (1975), Allentown (1997), and Altoona (2002). Steven Bourassa thoroughly details the land value taxation experience in the United S tates in Land Value Taxation, Theory, Evidence and Practice (2009) and this section borrows heavily from it. New York Beginning with the New York State (NY) value taxation was the shortest in lifespan. According to Re eb (1998), as cited in Bourassa (2009), t coupled with an overdue property assessment doomed the policy. The 1995 1996 assessment caused drastic changes in the tax bills and caused public opin ion to suggest that land was not being assessed accurately. As we will see later, this would

PAGE 33

33 not be the only time land value taxation was used as the scapegoat for insufficient assessment practices. Hawaii In Hawaii, land value taxation policy achieved it s goal of spurring development. However, it achieved this goal all too well in the eyes of much of the public who saw a hotel building boom as overdevelopment, particularly in Waikiki. Throug h a variety of legislature ma chinations from the mid 1970s to t he early 2000s property tax powers devolved to local levels and in many locales, rates on improvements actually were set higher than on land. In the cases of New York and Hawaii, we have seen land value taxation policies rescinded for failing to generate economic development by seemingly drastically changing tax bills (Amsterdam) and for succeeding in generating economic development too well thus seemingly drastically changing the character of a plac e (Waikiki) (Bourassa, 2009) The following description of Pennsylvania taxation experiment will detail why land value taxation persists in some areas of this state and why it has been rescinded in others. Pennsylvania Turning to the Commonwealth of Pennsylvania, we will now examine the trajectory of land value taxation policies in the Commonwealth which contains the three case study municipalities to be compared later: Altoona, Pittsburgh, and Wi lkes Barre. Understanding the history of property tax policy at the Commonwealth level will provide a baseline perspective from which to analyze the three case study municipalities. Pittsburgh was, from the beginning, at the center of the land value taxation movement. As the Gilded Age push ed the Steel City into economic boom times, land

PAGE 34

34 values within the c ity increased drastically. These i ncreases were largely attributable to the speculation of wealthy landowners keeping large amounts of land out of productive use. This situation was ripe for the followers of Henry George to seek legislative action, whi ch they would receive in 1913. As per legislation, tax rates on la nd would gradually increase while tax rates on improvements would decrease over a twelve year span. By 1925, this would culminate in a 1.95 tax rate on land and a 0.98 rate on improvements, a 2 1 ratio. Later on, a 1968 amen dment would allow the c ity to change the 2 1 ratio, which Pittsburgh and other municipalities wou ld ultimately take advantage of (Bourassa, 2009) 1951 Pennsylvania legislation permitted two rate property tax in the Comm Harrisburg, the Capital of the Commonwealth, being the first to adopt the policy in 1975. By this time, fervor for land value taxation policies in Pittsburgh and other municipalities around the Commonwealth had sw itched gears from discouraging speculation of large land holdings to promoting urban core revitalization. L egislation from 1993 and 1998 would open split rate property tax policies to school districts and boroughs, resulting in a total of 23 Pennsylvania taxing districts adopting the policy at some point as of 2008, including Altoona in 2002 (Bourassa, 2009) By 2001, the 88 year run of the two rate property tax in Pittsburgh would see its demise. As was seen in Amsterdam, NY in 1995, a long overdue asses sment cause tax bills to change drastically. In conjunction with a mayoral election year, the overdue assessment resulted in the two rate system becoming the scapegoat for tax bill complaints. The c ity does however still maintain an indirect split rate p roperty tax

PAGE 35

35 system. A myriad of tax abatements on new construction and renovations have been put in place result ing in a reduced effective tax rate on improvements relative to land. tablished a special assessmen t district known as the Downtown Pittsburgh Business Improvement Distric t, financed by a land based tax (Bourassa, 2009) Aside from Amsterdam, NY and Pittsburgh, PA, six other districts have rescinded the split rate tax system, all the result of perceiv ed injustices caused by overdue assessments. In Hazelton, PA, the policy never garnered broad support and was quickly overturned. In Coatesville, an overdue assessment dramatically shifted the tax burden onto residential property owners and away from busin esses and industries. Oil City abolished th e split rate system after faulty valuing of land and improvem ents in a 2003 assessment, adding to the perceived ineffectiveness of the policy in promoting economic revitalization. Bourassa (2009) provides a wealt h of information on the status of land value taxes in both Pennsylvania and New York. In the article, Bourassa provides details about the year the particular land value tax policy was adopted, the year it may have been rescinded, current rates on land, cu rrent rates on improvements, and the land to improvements tax rate ratios for each tax district As of 2008, sixteen districts maintain a split rate tax system. These municipalities and school districts have continued to employ the policy because it is be lieved that it encourages, or at least does not impede, economic development, that it is viewed as a more stable tax base, or that it is a more equitable and just tax.

PAGE 36

36 Of particular note in the article tax rate on land of 230.32 is str ikingly high. This, along with several of the other higher rates on land is the result of the use of land values from assessments that were not performed recently. As opposed to reassessing land values, Altoona has simply rais ed the tax rate on land value s (Bourassa, 2009) Previous S tudies of Land Value T axation While the universe of data for land value taxation research is limited, researchers have developed theoretical, general equilibrium, and regression models to help better understand the impact of t he tax regime. This section of the paper will draw upon John Land Value Taxation, Theory, Evidence and Practice (2009) edited by Richard F. Dye and Richard W. England to discuss the relevant literature su rrounding the effects of land value taxation. Theoretical Models First, theoretical models help establish a basis for analyzing real world empirical data that is found in general equilibrium and regression model that will be discussed shortly. Of particul aforementioned 1986 study that laid much of the ground work for modern analysis of land value taxation (Anderson, 2009) His study found that the move from a traditional property tax to a split rate t ax in which land is taxed at a higher rate than improvements generally increases land intensity. He also made the important observation that as a tax on improvements is lowered, land values increase. Conversely, as a tax on land is increased, land values decrease. These opposing forces and their magnitudes are crucial to understand the potential ra mifications of a split rate tax (Brueckner, 1986)

PAGE 37

37 Additionally, Bru e Analysis of the Effects of S In the case of a tax zone raising tax rates on land values and reducing tax rates on improvements to maintain equal yield to the previous regime Brueckner found that t he overall impact on land values depends on the relative si ze of the tax zone to which the tax regime applies. If the said zone is a small share of the marke t area, then housing prices would be exogenous and remain constant. This is because the area in which the tax regime applies is negligible compared to the m arket area in determining housing prices. While improvements per acre rise as a result of the lower tax rate on improvements Brueckner also notes that land value increase s as well This latter effect is the result of the reduction in deadweight loss asso ciated with the reduced im provement tax rate and increased land tax rate. A reduction in deadweight loss presents itself in the form of higher land rents, a residual of what revenue is left over after other factors of production earn their returns. The h igher land rents are captured by landowners. Deadweight loss under the old tax regime has become surplus value. Therefore, t his surplus value is reflected in higher land values. In this case, Brueckner concludes that the impact of these increased land r ents would be stronger than the impact of the direct tax on land. On the other hand, if the zone of the equal yield tax regime change encompasses the entire market area, then the value of land would likely decrease According to the li kely decrease in land prices is the result of the elasticity of housing demand the lack of locational advantage and the resulting reduction in the profitability of development In this case, the tax regime change does impact the price and supply of hous ing. The lowered tax rate on improvements means that

PAGE 38

38 improvements per acre are likely to increase In turn, this is likely to lead to an increase in the supply of housing and a decrease in price. Additionally, the increased taxes on land cannot be passed on to final demand as was described earlier in this chapter. Any attempt of the landowner to raise the price or rent of land would lead to less land being demanded, subsequently causing an excess supply of land and downward pressure on price (Oates & Schw ab, 2009) While the lower tax rate on improvements would still likely cause an increase in improvements per acre and deadweight loss would still be reduced, the s e effect s would be marginalize d by the drop in housing prices, leading to an over all reductio n in land values. chapter on the time of development. Anderson (1999) posits that a move from a traditional tax regime to a split rate system hastens timing of development and increases capital intensity, which means the amount of investment devoted to improving a land parcel. In Mills (1981a, 1981b, 1983) and Anderson (1986, 1993) conclude that property tax can alter the timing and capital intensity of development. However, Tideman (1982) offers an impo rtant grounding to the discussion. He clarifies that pure land value taxation is neutral with respects to timing of development. The aforementioned research modeled land value related to current land use, not the highest and best use. If land value is d etermined in a way that is unrelated to current land use, neutrality with respects to timing of development holds true. This is because c ost benefit analysis shows that both

PAGE 39

39 traditional property tax and land value tax subtract the same from the present val ue of both the choice to develop now and the choice to postpone development, assuming highest and best use is used to determine taxable value. effect on timing of development in this case is the result of differing as sumptions and frameworks. Anderson (1999) assumes a significant liquidity effect in his analysis, the result of increased costs associated with holding land whereas Tideman (1982) neglects this effect in his analysis in addition to focusing on the valuati on process of land. As a result, different conclusions about land Furthering the discussion of the impact of land value taxation on the timing of development, Arnott (2005) conducts research d that a tax on the raw value of land, similar to its highest and best use, is neutral in regards to timing of development. A residual land value tax, he finds, is non neutral in this respect as it is related to current land use, as was also pointed out by Tideman (1982). Arnott also concludes that there is a trade off here in attaining neutrality in timing of development. Post development raw land value would be highly complex and difficult to determine, likely resulting in unfair and arbitrary assessments. General Equilibrium Models General equilibrium mo dels are more ambitious in terms of the scope of assumptions made by the researcher. As Anderson (2009) describes, models of this vein attempt to provide description of the equilibrium conditions in all markets at the same time. These markets include but are not limited to land, buildings and other

PAGE 40

40 improvements, labor and output. Such models are often made to replicate the initial conditions of an economy and then used to calculate changes that occur in the aftermath of an intervention. Grosskopf begins t he examination of land value taxation through general tax incidence framework to find that a tax regime change from traditional property tax to a split rate scheme would r esult in increased equil ibrium land prices (Grosskopf, 1981) This was significant in establishing land value taxation as a tax policy switch that could potentially pay for itself. (Anderson, 2009) DiMasi (1987) developed a model of the City of Boston, MA and concluded that a switch to a land value tax policy would decrease land rents, increase improvements per acre, decrease housing prices, increase population density, reduce the area covered by urbanization, and increase wages. More modern analysis in thi s realm has been relatively scarce, although Nechyba (1998) and Haughwout (2004) are of note. Nechyba examined the effects of an increase in land tax and a decrease in capital tax with his model, which was set up to represent U. S. state and local governm ents. He concluded that this intervention would increase capital stock and decrease land values. Haughwout (2004) developed a model general corporate taxes while only retain ing the tax on land at its current rate. This Henry Georgian study concluded that private output would increase, land values would increase, private capital stock would increase, and population would increase but the provision of public goods and per capi ta tax revenue would decrease by over 50%.

PAGE 41

41 Ultimately, the differing conclusions reached in each of the general equilibrium studies are the result of differing model frameworks and assumptions. Grosskopf (1981) and Nechyba (1998) set up less location spec ific models and incorporated differing assumptions in building their general equilibrium models. DiMasi (1987) and Haughwout (2004) chose to develop models of specific cities; Boston and New York City, respectively. Inherently, the selection of different jurisdictions will likely mean the selection of different assumptions when building general equilibrium models. Regression Models An even more quantitative approach to examining the effects of a switch to land value taxation is regression analysis. By us ing this method, impact of the policy intervention on key factors of interest, known as dependent variables, can be isolated by the researcher. At the same time, the research can control for a number of other factors, known as in dependent or control variab les, which may also affect the dependent variable. (Anderson, 2009) In this way, the research can quantitatively examine the significance of one variable in relation to another. Studies by Mathis and Zech in 1982 and 1983 formed the groundwork for the regr ession analysis of land value taxation. In these papers, they studied the value of construction in 27 Pennsylvania cities, towns, and boroughs over the period 1976 1978. They computed two tax measures to make data in each municipality compatible and fit into single equation models. These measures were the ratio of the city tax rate on land to the city tax rate on improvements in 1977 and the same land to improvements ratio but from the combined city and county tax rates. They concluded that neither tax measure had significant impact on median or mean value of construction across t he Pennsylvania municipalities (Mathis & Zech, 1982, 1983) Also of note however, was

PAGE 42

42 the fact that only three of the 27 Pennsylvania municipalities studied by Mathis and Zech had implemented land tax rates that exceeded improvement tax rates at the time. Thus, there was not much variation among the tax measures across municipalities, possibly accounti ng for the lack of significance (Anderson, 2009) Bourassa (1990) was the fir st to use regression analysis to analyze the value of new residential building permits in the broader scope of the effects of land value taxation. To do this, he examined housing development in Pittsburgh, McKeesport, and New Castle, PA. He tested for l iquidity effect (i.e. the hastening of development as a result of a higher land tax rate) and for incentive effect (i.e. the encouragement of further housing development as a result of lower relative tax rates on improvements.) He quantitatively found tha t land value taxation may have impacted residential housing permits in Pittsburgh, but not in McKeesport or New Castle. More specifically, Pittsburgh housing development exhibited incentive effect but no liquidity effect. Oates and Schwab (1997) explored new building activity in 15 cities in the multistate region of the Pittsburgh metropolitan area using time series data. Conducting before and after analysis, they conclude that after 1979 1980, the period when land to improvement tax ratios shifted in Pit tsburgh, something dramatic happened to building activity in Pittsburgh compared to the other cities in the region. The increased building activity, and therefore increased value of building permits, was caused by both the change in tax regime and the occu pancy rate, results indicated. Additionally, the effects attributable to the tax regime change are muddled by the influence of the property tax that a land tax did not cause a building boom in

PAGE 43

43 Pittsburgh, but it did allow the city government to avoid policies tha t might have undercut that boom (Schwab & Harris, 1997 [as cited in Anderson, 2009]) Staying in the Steel City, Pollakowski (1982) exa mined property transactions in Pittsburgh from 1977 1981. As previously mentioned, this was a period of great change to improvement tax ratio was 2:1. In 1979 it was increased t o 4:1 and then increased to parcel being transacted. He would go on to find a positive and discernable effect of the land tax rate. Of the 6812 properties transacted in 1979, Pollakowski estimated that 60 were attributable to the land tax increase. Thus, although the effect of the change in land tax rate was discernible and significant, its overall magnitude was very modest. Tideman (2000) examined building permit data in Pennsylvania cities from 1980 1994 to determine impact on building activity. Their dependent variable was defined as the number of building permits per person per month in each city. They then developed an i ndependent variable which improvements then divided that difference by the average ratio of assessed value to sales value in the city. In this way, they would eliminate variatio n based on differences in assessment ratios across cities. They found that an increase in this differential equated to a significant increase in building permits. Similarly but broader in research scope, Tideman and Johnson (1995) studied building permit data in Pennsylvania cities over the period 1980 1992 to determine if a

PAGE 44

44 shift of taxes from improvements to land produced economic growth. However, they went on to find that available data are insufficient to identify any effect adequately. Comparison App roach Finally, after numerous of other studies of his own, Cord (1987) used a comparison method to examine construction activity in Pennsylvania cities that shifted to split rate taxes. In this study, Cord compared such cities to neighboring cities that c ontinued to employ traditional property tax. While the study found that cities with the split rate tax experienced more construction activity, the method did not control for other factors that may have impacted building activity nor di d it account for sel ection bias (Anderson, 2009) Summary of Previous Studies This section has elaborated on previous studies conducted on land value taxation using varying models. Overall, it appears clear that a shift from traditional property tax to land value taxation sh ould lead to increases in improvements per acre and land intensity. Expectations about the effect on timing of development and land values are less clear, however. Studies illustrated in this chapter used different frameworks in their models and in some cases made different assumptions about such things as the elasticity of housing demand and the significance of the liquidity and incentive effects. As a result, definitive expectations about the effects of land value taxation on timing of development, lan d value, and other variables remain elusive.

PAGE 45

45 Relevance to Planning Economists rave about the prospects of economic efficiency emanating from land value taxation. As far back as David Ricardo, they have long considered a tax on land to be the most ideal tax because the return on holding land is nothing but economic surplus, not the consequences of economic actions by landowners (Netzer, 1998) However, there is more to this notion than just economic efficiency. As Tideman (1998) ion can be generalized to the principle that people should pay for all of their appropriations of natural opportunities according to the opportunity costs of those appropriations, and that the resulting revenue should be shared equally. There are importan t applications of this principle to questions of environmental protection, relieving congestion, efficiency resource use, population growth and general economic Tideman is correct and planners know this as well. Much of what the field of urban pl anning consists of is about the efficiency of cities and regions to enhance the lives of citizens. Planning goes beyond economic development. While economic concerns are often at the crux of land value taxation discussion, let us not forget the very title of Progress and Poverty comprehensive than mere economic development. He set out to address the paradox of persistent poverty and inequality in a world of constant technological advance that should enrich the happiness and welfare of all members of society, improv ing the overall human condition (American Journal of Economics and Sociology, 2005) Property tax policy influences the physical, econ omic, and societal structure of settlements everywhere. Planners must be aware of t he ramifications of such policy. Not only is property tax a primary source of revenue for local governments and thus of

PAGE 46

46 municipal planning departments, but it also influences development patterns, which effect everything from t raffic patterns to infrastructure layout to public services provision to growth management. It also has an impact on income and wealth distribution and environmental protection by way of urban sprawl development. As a result, land value taxation and prop erty tax policy in general should not be taken lightly or dismissed as merely an economic development tool by planners.

PAGE 47

47 Figure 2 1. Tax e ffects on an u pward s loping s upply [Recreated by author from Oates, In Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. England. Cambridge, MA: Lincoln Institute of Land Policy: 51 71.] Figure 2 2. Tax e ffects on a f ixed s upply [Recreated by author fro m Oates, W. E. & Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. England. Cambridge, MA: Lincoln Institute of Land Policy: 51 71.]

PAGE 48

4 8 CHAPTER 3 ME THODOLOGY In this chapter, the process from which the analysis presented in this paper is derived will be explained. In particular, the examination of the methodology used will include the study approach, study areas, data collection and limitations. The chapter preparation for the analysis, findings, and discussion of Chapters 4 & 5. Study Approach The literature review from the previous chapter provided background informati on concerning the debates around land value taxation and its relationship to various aspects to housing, development, and income. Based on the arguments made previously, the presence of land value taxation should raise income to a higher level than in pla ces without it, all other things being held equal. The reason for this is simply that it allows the market to behave in a more economically efficien t way, reducing deadweight loss (Brueckner, 1986) Arguments also indicate that a switch to split rate tax in which the millage rate on land is higher than that on improvements leads to in crease in improvements per area (Anderson, 2009) From these two hypotheses, theoretical expectations for a set of income and housing variables can be derived, which is what t his thesis sets out to examine. This research seeks to understand the differences between the theoretical effects of land value taxation and tangible observations from United States ( U S ) Census and American Community Survey (ACS) data. The work encompas ses three case studies, all from the Commonwealth of Pennsylvania: Altoona, Pittsburgh, and Wilkes Barre. Using an analytical approach, cross sectional before an after analysis will be conducted

PAGE 49

49 on the three cases study areas to better understand how inco me and housing characteristics were changed by the policy intervention of land value taxation. Originally, the goal of the research was examine changes in property values following the adoption of a split rate tax in which land was taxed at a higher rate than improvements as a property tax policy. This would provide insight into the viability of a property tax regime by indicating changes in demand for property within the tax jurisdiction. However, as researched progressed it became clear that the objecti ve would need to be adjusted because of data collection limitations These changes in methodology as well as data collection and limitations are illustrated in the rest of this chapter. Case S tudies Beginning with an eye toward assessing the effectiveness of land value taxation, research began by looking at instances of adopted land value taxation policies. These instances in the United States are described in Chapter 2. Attempting to grasp the effects of the tax regime in the most contemporary of context s, the case of Altoona was first to be studied, due to its recent adoption of the heavier tax rate on land than on improvements. Additionally, research would expand to cover two other municipalities for the purpose of comparison, both in Pennsylvania. Pi ttsburgh was chosen due to its recent rescinding of city wide land value taxation. Wilkes Barre was chosen as a control case study, similar in demographics and industrial makeup to both Altoona and Pittsburgh but never having adopted land value taxation. The three case study methodology was very important to this research. Using three case studies allows the research to account for many extraneous influences that would not be covered with only one study area. For example, a reduction in income

PAGE 50

50 growth obs erved for one case study may indicate different conclusions if compared to multiple case studies. Individually, a reduction in income growth for a study area may seem like a negative outcome. However, if paired with other case studies that show greater r eductions in income growth, suddenly the outcome from the original case study does not appear as negative. More specifically, case studies can be chosen to cover the recent removal of the policy intervention and its effects as well as a control case study which can yield effects in the event of non implementation of the intervention. In this research, alternative case studies were selected to allow the examination of a recent removal of land value taxation policy and examination of an area that has never adopted land value taxation. In particular, case studies of the City of Pittsburgh, PA and the City of Wilkes Barre, PA were chosen as alternatives to the City of Altoona, PA for reasons listed above. First, utilizing case studies from within the same co untry and within the same state or, in this case, commonwealth, was crucial to mitigating variation produced by differing federal and state or commonwealth legislation. Because Pittsburgh and Wilkes Barre are, like Altoona, within the United States and wi thin Pennsylvania, all three jurisdictions would likely fall under the same federal and state or commonwealth law. As a result, observed differences in income or housing characteristics would not be induced by differences in federal or commonwealth legisl ation. In an effort to further mitigate the effects of extraneous circumstances, the three selected case studies share similar demographics and industrial profiles. Like much of Pennsylvania, Altoona, Pittsburgh, and Wilkes Barre are all cities that ros e to significance by way of the manufacturing industry. Manufacturing was the abundant

PAGE 51

51 source of economic might for th e factories of the Commonwealth (Stevens, 1955) Altoona and Wilkes Barre have been similar in economic reliance on manufacturing, with b oth cities employing roughly 19% of its workforce in the sect or in 2000, according to Table A 1 Employment Figures by Industry. Pittsburgh is also similar in industrial makeup to the other two cities, albeit at a larger scale. Additionally, while Altoona recently adopted a land value taxation policy in 2002, Pittsburgh rescinded its policy after employing it since 1913. By examining the case of Pittsburgh, the effects of removing a land value taxation policy can be studied in contrast to the Altoona case Study A reas The following are brief descriptions of the case study areas of Altoona, Pittsburgh, and Wilk es Barre. Additionally, Table A 1 Employment Figures by Industry in Appendix A and overall industrial diversity. Altoona The City of Altoona is located in central Pennsylvania with a population of 46,320 (2010 U S Census) A former railroad hub between Philadelphia and Pittsburgh, Altoona has seen economic and population decline si nce their heights in the mid 20 th century but has seen resurgence s over the last 20 years. The c ity adopted a land value taxation policy in 2002. 2010 Bureau of Labor Statistics data reveals that Altoona is reliant on the manufacturing, retail trade, and health care and social services industries for over 53% of its employment. These sectors are also the three most prevalent sectors in the national economy, only amounting to roughly 40% of total employment nationally, however.

PAGE 52

52 Between 2001 and 2010 Altoon a gained over 1250 jobs relative to its share of national employment despite losing 143 total jobs overall. This means that the economy of Altoona shrank over the period but not as much as that of the national economy Competitive shift figures show that Altoona was a highly competitive environment for the manufacturing, retail trade, and administrative and waste services industries. Each of these industries grew more rapidly in the local area than nationwide. Altoona also remains strong in the transpor tation and warehousing sector. Historically a railroad town, Altoona may capitalize on rising fuel costs and once again utilize this industry to catalyze future growth across a myriad of industries within the metropolitan area. Pittsburgh The City of Pit t sburgh is located in western Pennsylvania with a population of 305,704 (2010 U S Census) A former powerhouse in the steel industry until the latter 20 th century, Pittsburgh shares the industrial heritages of Altoona and Wilkes Barre albeit at a much lar ger scale. The Steel City adopted a land value taxation policy in 1913 but reverted to a traditional property tax in 2001. stalwart manufacturing employment in 2010 was less than the national average. Manufacturing, retail trade, and health care and social assistance are also the most significant in Pittsburgh, but account for just below 42% of employment. The industrial mix of Pittsburgh was a positive factor in job growth, by itself creating approximately 24,000 jobs. Industries such as health care, education, and accommodation and food services, in fact growing significantly in Pittsburgh, grew nationally. Additionally, between 2001 and 2010, shift share analysis shows that

PAGE 53

53 competitive shi ft was extraordinarily negative resulting in a loss of over 45,000 jobs in total. This was particularly the case in the health care and social assistance sector, whose competitive shift amounted to over 14,000 alone. Evidently, that sector in Pittsburgh was significantly outperformed nationally. Wilkes Barre The City of Wilkes Barre is located in northeast ern Pennsylvania w ith a population of 41,498 (2010 U S Census) driven city, Wilkes Barre is of similar size and economic base to Altoona with the notable exception of its proximity to a 75,000 resident city in Scranton. Wilkes Barre has never implemented land value taxation. Wilkes Altoo na and Pittsburgh in terms of reliance on manufacturing and industrial diversity. Manufacturing accounts for just under 13% of employment in Wilkes Barre and manufacturing, retail trade, and health care and social assistance constitute just below 47% of t he total. Like Altoona and Pittsburgh, Wilkes Barre has a strong mix component according to shift share analysis. By itself, the mix contributed a gain of 250 jobs over the period. Health care, education, and accommodation and food services were all secto rs that showed high growth nationally and also held significant employment in Wilkes Barre. Competitive shift analysis shows local strengths particularly in transportation and warehousing, management of companies and enterprises, administrative and waste services, and arts, entertainment, and recreation. However, several other industries such as manufacturing, health care, and educational services, performed below national

PAGE 54

54 standards, resulting in significant local jobs losses amounting to nearly 2400 due to lack of local competitiveness. Cross Sectional Study Cross sectional analysis will be used in this research. By examining data from a specific point in time in all three study areas, an overall picture can be obtained (Kumar, 2005) The focus of the cr oss section for this study was originally property value but was adjusted to selected income and housing characteristics over time. The research first attempted to gather assessed property values for the three case study areas from a central data clearing house at the state level. However, when contacted, the Commonwealth of Pennsylvania indicated that such data was kept at the local level. From there, gathering comparable data became increasingly difficult as local municipalities kept their property tax rolls in various locations and with various subcontracted firms. In the end, time and financial constraints terminated the effort to focus cross sectional analysis on property values. Research focus then turned to the readily available housing and income characteristic data in U S Census and American Community Survey datasets. Cross sectional analysis remained the method used for this research. This type of data analysis was chosen because of its simplicity of design and as a component of before and afte r analysis. Before And After Study Design In order to ascertain the effects of land value taxation over the study areas, this research will be developed with before and after analysis. Utilizing two sets of cross sectional data collection points from the same case study areas, before and after analysis allows the research to determine changes in phenomenon or varia bles between two points in time (Kumar, 2005) Figure 3 1 Before and After Study Design

PAGE 55

55 and Figure 3 2 Measurement of Change Through a Before and After Design depict how this methodology works. In Figure 3 1 the general conceptual framework of before and after design is portrayed with two measurements of the same population surrounding a program or intervention on a ti meline. Figure 3 2 illu strates the potential measurements of change in the before and after study design. The before and after study design ideally fits this research. Altoona adopted land value taxation in 2002, Pittsburgh rescinded it in 2001, and Wilkes Barre has never adopt ed it. Data for decennial censuses is produced each decade, which was originally thought to be ideal for this research. The cross sections of time were to be 2000 and 2010, both surrounding the interventions of policy adoption in Altoona and policy remov al in Pittsburgh. While cross sections were ultimately adjusted to 1990, 2000, and 2009, the before and after study design remains the method of choice for this research. This method is not unheard of in planning research. Specifically pertaining to the and after design to examine new building activity in Pittsburgh and surrounding municipalities Sc hwab (1997) article thesis. Data Collection U S Census of Population and Housing and American Community Survey The original intent in collecting data for this research was to use county pr operty assessor tax rolls from years before and after the policy interventions to examine changes in assessed property value. As previously mentioned, the research was

PAGE 56

56 unable to incorporate pro perty assessor tax rolls from a statewide database or other ce ntralized location In attempting to obtain the tax rolls from individual counties, the author was either redirected to private consultants contracted for assessment services or told that archived data from previous years could not be found. Additionally the current tax rolls that were available from county assessors carried with them usage fees that were beyond the budget of the research Undeterred, the study was adjusted to incorporate more readily available U S Decennial Census and American Communit y Survey (ACS) data. The U S Census of Population and Housing produces data on a decennial basis. Ideally, this research would have used data from the 1990, 2000, and 2010 U S Ce ns us es of Population and Housing. However, at the writing of this thesis, housing and income characteristic data for Census Barre, had not been released for the 2010 Census. As a result, utilization of the 2010 data was unavailable as well. This research then sought out ACS data. While containing higher margins of error than data from decennial censuses data from the ACS uses the same geographic boundaries as decennial censuses and, most importantly, is produced annually. Accordingly, the study was adju sted to examine income and housing characteristic data sets from the 1990 and 2000 U S Cen su ses on Population and Housing in conjunction with like data sets from the 2005 2009 ACS. The research examines a total of ten variables from the data sets, three i ncome characteristics and seven housing characteristics. While analysis of the various housing

PAGE 57

57 and income characteristics of the three study areas may yield overall impacts of land value taxation, the variables are related to the tax at varying levels. Generally, housing characteristics are more closely related to land value taxation tax rate on improvements relative to land, which theoretically incentivizes construct ion versus the purchase of land. Income characteristics, meanwhile, are somewhat removed from the direct impact of land value taxation. Income figures face influence sw itch to a land value tax would necessarily affect housing and other markets as intermediaries between the policy and income. In this way, housing characteristics have more of a primarily relationship to land value taxation whereas the relationship to inco me characteristics can be considered secondary. Individually, housing figures also vary in their relation to land value taxation. In theory, density measures and units in structure bear the closest interaction with the tax policy, as the tax simultaneousl y raises the appeal of improvements and diminishes the appeal of land. Variables such as occupancy status, property value and gross rent are further removed from direct relation to land value tax. Total housing units and the age of housing stock (year st ructure built) may be directly impacted by the tax policy depending on existing supply and demand conditions for housing. These variables are listed i n Table 3 1. List of Variables Tested in Before and After Model. Hypothese s of expected results for each variable are dependent upon both the date of the cross section and the particular study area. Theoretical expectations of each variable will be detailed in Chapter 4 Results and Analysis.

PAGE 58

58 Bureau of Labor Statistics Data In order to better address extran eous influences to the U S Census and ACS data, data sets from the Bureau of Labor Statistics (BLS) were brought into the research to better describe the economic landscapes of the three case study areas. Due to data availability constraints, BLS data se ts from the years of 2001 and 2010 for each study area were used. The results of this research were used as economic context enhancements to supplement the variable analysis in Chapter 4. The economic profiles of the three case study areas produced by th is BLS research can be foun d in Appendix A Data Review & Economic Profiles. Utilizing these data sets, a tr io of analytical techniques was composed. The descriptive analysis of location quotients offers a snapshot of the states of each economy at particu lar cross sections of time. From a more dynamic and predictive perspective, shift share analysis and Esteban Marquillas Extension attempt to illustrate changes in the industrial m akeup of each economy over time (Blakely & Leigh, 2009) A location quotien t is the ratio of the percentage regional employment in a particular industry to the comparable percentage in a benchmark area. A location quotient equal to 1 indicates that the area has the same percentage of employment in that industry as does the natio n. Location quotients greater than 1 mean that the area a higher concentration of employment in that industry than does the nation and they the sector likely an export industry locally. By extension, a location quotient of less than 1 likely belongs to a local import industry and that the area has a less than proportionate share of employment in the industry (Blair & Carroll, 2009) (Blakely & Leigh, 2009) Shift share analysis allows for the conducting of dynamic analysis to inform decision makers of ch anges in the local economy. This type of analysis disaggregates

PAGE 59

59 growth into economic growth, mix component and competitive component. Economic growth is an indication of how the local economy has grown or shrunk with growth or shrinkage in the overall ec onomy. Mix component depicts how well the diversity and mix of industries locally is performing relative to the mix of industries in the overall economy. Competitive component measures how the performance of a particular local industry relative to how th e particular indust ry is performing in the overall (Blair & Carroll, 2009) (Blakely & Leigh, 2009) (Edwards, 2007) As its name suggests, Esteban Marquillas Extension merely expands, upon Shift share analysis. This technique more clearly identifies a loca component by redefining competitive shift and adding a fourth effect known as allocation effect (Edwards, 2007). The allocation effect creates a method for properly distributing credit for local economic growth (or decline) in an i ndustry among overall economic growth, industrial mix, and competitive effects. These alterations allow for a more accurate portrayal of the attractiveness of a particular local economy (McDonough & Sihag, 1991). This research employs these analysis techn iques in order to better illustrate where the economies of the three study areas have been and where they are potentially being driven. With this analysis in mind, the research is in better position to determine what changes are the results of the effects of land value taxation and what changes are from extraneous origins such as unassociated changes in the local or national economy Limitations Several facets of this work have their limitations. These limitations begin with time and access concerns then also cover the before and after study design, case and

PAGE 60

60 variable selection bias, data collection processes and decisions, the general qualitative approach and conclusions, and even the attempts to mitigate pre existing limitations. Time and access were con straints that hampered this research from the beginning. With more time and access, either through increased scope of research time or increased financial wherewithal, the original intent of this research may have been realized. Property assessor tax rol ls are available to the public under Right to Know legislation and are surely archived somewhere, either digitally or in physical form. were more time and financial resourc es allotted. Despite the advantages of the before and after study design, limitations to it exist and inhibit this research. First, before and after design only measures total change in variables and does not separate the effects of extraneous variables ( Kumar, 2005) The research attempts to mitigate this limitation by selecting case study locations that avoid differences in state and federal legislation and differences in economic composition. Economic differences between case study areas are further mi tigated by the incorporation of Bureau of Labor Statistics data in location quotient, shift share analysis, and Esteban Marquillas Extension tables. Secondly, before and after analysis is also limited by potential regression to the mean. This means that t he design does not account for the tendency of extreme observations in the data before the policy intervention return to the average, with or without the effects of the intervention (Kumar, 2005) The research attempts to control this limitation by includi ng variable data from the 1990 U S Census of Population and Housing. This data was collected ten years

PAGE 61

61 prior to the observations of the year 2000 and should therefore indicate the presence of extremes in the data. The possibility of selection bias in cas e and variable selection also limits the research. Attempts to reduce case selection bias included choosing all study areas from the same state or, as is the instance in this research, commonwealth. All three study areas fall under similar state and fede ral legislation. The implementation of a control case, the City of Wilkes Barre which has never adopted land value taxation, was also conducted with the goal of reducing case selection bias. Variables were selected by the author as indicators that are si gnificant to land value taxation. These selections were based on land value taxation literature but still may not include all significant characteristics of municipalities that are affected by a change in property tax regime. Data collection processes als o were performed with limitations. As previously mentioned, the inconsistency of comparing 1990 to 2000 data with 2000 to 2009 data was an unfortunate byproduct of time constraints. If housing and income characteristic data for Census r om the 2010 U S Census of Population and Housing were the limitations imposed by this inconsistency. Furthermore, the usage of ACS data also contains greater margin of error than decennial U S Census data, introducing limitations in drawing conclusions a n d d e t e r m i n a t i o n s from changes in data between 2000 and 2009. While informative as this qualitative research method is, it is nevertheless limited by its lack of quantitative analysis and concl usions. In order to mitigate this, a simple ordinary least squares statistical regression was attempted. However, time and data

PAGE 62

62 constrained this portion of the research as shape file geographies from the U S Census data clearinghouse did not correlate to 2005 2009 ACS data for Census defined mismatch and time to do so was unavailable to the research. Additionally and ironically, attempts to mitigate the previously stated limitations by incorporating BLS data analysis were limited by a few factors. First, labor statistics for each individual study area were unavailable. Instead, the research was faced with using metropolitan statistical area (MSA) or county level data whi ch encompass each study area MSA level data was chosen because such areas are generally focused around central cities such as Altoona, Pittsburgh, and Wilkes Barre 1 Therefore, MSA level data was selected for its ability to more closely resemble the loc al economies of each study area. Despite the similarity of the MSAs with their respective study areas, t his analysis is still limited because the BLS data includes observations from across the MSAs, not just the study areas in question. For instance, labo r statistics for the Scranton Wilkes Barre MSA were used as a proxy for labor statistics in the City of Wilkes Barre. Secondly, archived data from before 2001 was not immediately available from the BLS. As a result, data sets from 1990 and 2000 which cou ld have corresponded with the U S Census data sets was not used. Instead, 2001 and 2010 data sets from the BLS were used in an effort to maintain cross sections before and after policy interventions and over approximately a ten year span. 1 Central city designation in Wilkes

PAGE 63

63 Summary This wo rk analyzed changes in income and housing characteristics that occurred as results of the adoption, removal, or neglect of land value taxation policy by using before and after study design. Before and after study design uses two sets of cross sectional da ta collection points, each surrounding a policy intervention in time, to determine changes in a phenomenon or variables between two points in time. The City of Altoona was chosen to depict changes after the adoption of land value taxation policy while Pit tsburgh was chosen to show changes after the removal of the policy and Wilkes Barre was selected as a control case study because it has never implemented the policy. The three study areas were analyzed to determine how theoretical effects of land value tax ation on selected income and housing characteristics manifest themselves in U S Census and ACS data.

PAGE 64

64 Figure 3 1. Before and a fter s tudy d esign [Source: Kumar, R. (2005). Research Methodology: A Step by Step Guide for Beginners. Second Edition. Thousa nd Oaks, CA: Sage Publications.] Figure 3 2. Measurement of c hange through a b efore and a fter d esign [Source: Kumar, R. (2005). Research Methodology: A Step by Step Guide for Beginners. Second Edition. Thousand Oaks, CA: Sage Publications.]

PAGE 65

65 Table 3 1 List of variables tested in before and after m odel Variable Category Household Income Income Per Capita Income Income Income to Poverty Ratio Income Housing Units Housing Occupancy Status Housing Units in Structure Housing Gross Rents Housing Va lue of Owner Occupied Housing Units Housing Year Built Housing Densities of Housing Units and Population Housing

PAGE 66

66 C HAPTER 4 RESULTS AND ANALYSIS O v e r v i e w Having now described data from selected income and housing variables and analyzed industry p rofiles of the three case study areas, observations provided by the U nited S tates (U.S.) Census and American Community Survey ( ACS ) data tables will be directly compared to land value taxation theory. Beginning with housing variables and continuing with i ncome variables, expectations of each variable based on theory will be discussed followed by observed data and conclusions derived from the two i n a d e t e r m i n a t i o n s e c t i o n This categorical analysis will provide a clear synopsis of the presence of land value taxation in U S Census and ACS data. Housing Variables This section will compare theoretically expectations with observed results for housing variables. There are a total of seven housing variables in this study. Variable Total Housing Units Expectations based on theory Land value taxation theory pertaining to housing unit construction is more pertinent to density measures. A change to this type of property tax would likely produce a higher density of housing unit construction. Because the tax on land would result in higher costs for land and the relative reduction in the price of improvements, urban areas would be expected to experience increased floor to area ratio. For instance, detached single family units may become less affordable of a housing option than townhouses b ased on land total land coverage. Pertaining to total housing units, supply and demand would have a much greater impact than the property tax policy. Land value

PAGE 67

67 taxation would merely guide the concentration of such units, likely toward increased density. However, because the three urban areas in this study are essentially built out older cities, the decreased costs for improvements may prompt an increase in overall units where land value taxation is in place. The combination of higher costs for single uni t, detached homes and reduced costs for multi unit housing could increase the overall amount of units. Additionally, according to Bourassa (1990), land value taxation increases the holding costs of vacant parcels and reduces the costs to develop improvem ents on parcels, which Bourassa calls the liquidity effect and the incentive effect, respectively. Both of these effects should theoretically increase overall unit development in municipalities using land value taxation. It would be expected that in 2000, degree of housing unit growth in all three cases due to its status as the sole employer of land value taxation at the time. In 2009, we would expect Altoona to have the greatest percentage gain in overall housi ng units of the three case studies as a result of its adoption of land value taxation and since Pittsburgh rescinded it in 2002. Observed Data from Table 4 1 Total Housing Units indicates that in 2000, all three cases studies had experienced a decrease in total housing units over the previous ten years. Wilkes Barre lost 2.12% of total units, Pittsburgh lost 3.99%, and Altoona lost 4.48%. During the 2000 Wilkes housing unit total again fell by 1.53%

PAGE 68

68 Determination In the case of this variable, theoretical results were not evident in observations from either 2000 or 2009. The only evidence possibly pointing to the significance of l and value taxation is that while total units continued to fall in Altoona, reduction slowed from 4.48% in 2000 to 1.53% in 2009. Pittsburgh continued decline in its amount of total housing stock during the 1990s is likely due to the declining role of its e conomy in regional and national markets. Its rebound during the first decade of the 20 th century could be attributable to growing employment in the health care industry, implementation of property tax abatements on new construction, and a national housing construction boom in the first half of the decade. The property tax abatements are of particular interest in that they ha ve essentially replaced the results of land value taxation by reducing the effective tax rate on improvement s relative to the land ta x rate (Bourassa, 2009) In examining Altoona continued reduction in housing units, it becomes difficult to understand why the pattern continued from 1990 through 2009. First, the adoption of land value taxation should increase the development of housing units. Second, the national housing boom of the early 2000s should have impacted Altoona like it did Pittsburgh. Third, Altoona retained more employment between 2001 and 2010 than Pittsburgh or Wilkes Barre. Yet still, Altoona experienced a decrease in ho using units. A possible explanation for this is the move from blue collar to white collar employment in the Altoona e conomy in conjunction with the c employment generally leads to less dense housing development patterns. Unable to find housing of sufficient acreage within the city limits, new white collar employees may

PAGE 69

69 hav e chosen to reside outside the c ity thus decreasing the deman d for housing units within the c ity. Variable Occupancy Status Expectations ba sed on t heory Much of the theory pertaining to total housing unit construction als o applies to occupancy status. Pertaining to occupancy status again supply and demand would have a much greater impact than the property tax policy. I quidity and incentive effect hypotheses hold true, excess units may be produced in the pursuit of revenue to pay for holding costs leading to decreased occupancy rate. Assuming there is an impact from liquidity and incentive effects; by 2009 we would expe ct occupancy rate decreases in Altoona, increases in Pittsburgh, and a constant rate in Wilkes Barre. Observed According to Table 4 2 Occupancy Status, occupancy increase by 1.39% in the 1990s, and fell in Pittsburgh and Wilkes Barre by 2.21% and 5.31%, re spectively. Between 2000 and 2009, the occupied to total unit ratio fell in all three study areas. The number fell by just over 2% in Altoona, by 2.35% in Wilkes Barre, and by just over 4% in Pittsburgh. Determination 2009 figures may reflect land value taxa tion theory a ssuming impact from liquidity and incentive effects and oth er assumptions about the ability of white collar workers to find palatable housing within city limits Altoona experienced a decrease in occupancy rate over the 2000s after an increa se during the 1990s. Pittsburgh saw its occupancy

PAGE 70

70 status decline at an increasing rate. These are both changes that could indicate the presence of land value taxation. drop in total housing units. As manufacturing and other blue collar jobs continue to decline while health care, education, and other white collar employment continue to grow, units that were the former residences of lower income workers may remain vacant. Ne w higher wage workers locate beyond the c ity limits and can afford to commute into the c ity for work. Thus, as the liquidity and incentive effects induce landowners to maintain more units within the jurisdiction of the land value tax regime, fewer workers might seek housing within the jurisdiction leading to a decline in occupancy figures. Variable Units in Structure Expectations based on t heory taxation shifts the burde n of property tax from an equal distribution on both land and improvements to an increased burden on land versus improvements. As a result we would expect an increase in improvements per area of land. Thus an increase of the amount of units in structure s is likely to occur in municipalities adopting land value taxation. Accordingly, by 2009 we would expect to see an increase in units in structure composition in Altoona, a decrease in Pittsburgh, and no change in Wilkes Barre. At the very least, it would be expected that units in structure composition should decrease more in Pittsburgh than Altoona.

PAGE 71

71 Observed Between 1990 and 2000, Altoona saw little change in its units per structure makeup with the exceptions of the 50 unit or more structures category whi ch increased by 2.17%. Pittsburgh remained unchanged except that single unit detached structures increased by 2.64% and single unit attached structures decreased by 1.45%. Wilkes Barre also experienced an increase in single unit detached structures and a decrease in single unit attached structures in addition to an increase in two unit structures. Between 2000 and 2009, Altoona remained virtually unchanged with a slight increase in the amount of single unit detached homes. Pittsburgh also essentially mai ntained its housing unit makeup with only an increase in the amount of single unit attached structures. Wilkes volatility with an increase in single unit attached structures of nearly 5% and decreases Determination Little evidence suggesting an impact from land value t axation can be found in Table 4 3 unit detach homes is the o pposite of what would be expected theoretically. Reasoning for this lack of the expected effects of land value taxation on units in structure may also be tied to the shifting economies of Altoona and Pittsburgh. Higher wage earners tend to demand single unit housing, resulting in lower concentrations of little effect on units in structure composition. Despite the removal of official land value taxation policy in 2001, of the tax regime by effectively reducing the tax rate on improvements relative to the tax

PAGE 72

72 rate on land. The abatements should theoretically encourage more improvements to be consumed and less land, causing increases in the viability of multi unit structures. increased unit to structure ratios, assuming all other factors remained unchanged. Variable Property Val ue Expectations based on t heory The relationship between land value taxation and property values is complicated. According to Brueckner (1986), if adoption occurs in a relatively small portion of the market area, housing prices are likely to remain consta nt with the greater market at large. Combining constant housing prices with increased improvements per area and the reduction in deadweight loss caused by the shifted to land value taxation will produce an increase in property value. In this case, the be nefits of the reduction in deadweight loss will outweigh the value depressing effect of the increased land tax rate. In the case of an entire market area raising the tax rate on land and decreasing the tax rate on improvements, housing prices are likely to fall due to lack of locational advantage in the market and a reduction in the profitability of development. Additionally, the increased tax rate on land cannot be passed on to final demand. Therefore, in this scenario land values should fall. For more on Brueckner (1986), see Therefore, i f theory stays true to form, by 2009 Altoona shou ld experience a decrease in property values a s it adopted land value taxation in 2002 While small in land area compared other cities, Altoona represents the largest portion of its market area. Therefore, a change in property tax regime within the City of Altoona would have a significant effect on the housing market. Pittsburgh should experience an increase as

PAGE 73

73 it also represents the largest portion of its market area and rescinded land value taxation in 2001. Wilkes Barre should experience no change. Observed Between 1990 and 2000, owner occupied housing unit market composition in the three study areas moved simil arly. The housing unit market shifted as values for units rose from a predominantly below $35,000 to the $50,000 to $150,000 range. In particular, Altoona saw a drastic reduction in the amount of units valued at $25,000 or less. While all markets continu ed to move in the same increased price direction occupied housing unit market was clearly dichotomous with increases in high end units and decreases in low end categories. A ltoona also saw decreases in low end units and Increases in middle to high end units but growth in higher valued units was less vigorous than in Pittsburgh. The market for owner occupied housing units in Wilkes Barre shifted less than in Pittsburgh and Al toona. Determination D ata from Table 4 4 Value of Owner Occupied Housing Units for the year 2000 does not reflect theoretical effects of land value taxation. Differences between Altoona, which adopted land value taxation in 2002, and Pittsburgh, which resci nded the tax regime in 2001, were slight at best. Theoretically, the two municipalities would have experienced differences in the opposite direction. property values. Beca use the tax abatements effectively reduced the tax rate on improvements relative to that of land value, the Pittsburgh housing market witnessed a

PAGE 74

74 departure from land value taxation policy in name only. As a result, changes in property values in Pittsburgh are likely to be extraneous from the presence or absence of land value taxation policy. In Altoona, the lack of theoretical effects on property values manifesting themselves could be attributable to the stable economy of Altoona relative to the region and the nation between 2001 and 2010. By increasing its share of employment relative to the nation and attracting higher wage employment, property investment may be likely to also increase relative to the region a nd nation, especially considering the c recently introduced lower relative tax rate on improvements. Increased property investment could spell increased property values, thus negating the theoretical effects of property tax regime change. share of the regional housing market was large enough to impact housing prices may have been incorrect. If so, locational advantage and gains in land rents associated with deadweight loss reduction may have been enough to lead to increased property levels. Variable Gross Rent Expectations b ased on t heory The theoretical impact of land value taxation on gross rents is also complex. Assuming land value is a proxy for gross rents, theoretical effects of land value taxation on gross rents should mimic its e ffects on land value. The effect of the adoption of land value taxation on gross rents is likely to depend on the geographic and market scope of adoption. rescinding lan d value taxation in 2001, as the Steel City represents a significant portion of the market area. With Altoona adopting the policy in 2002, that city would likely see a

PAGE 75

75 decrease in gross rents as it also represents the most significant portion of its marke t area. Wilkes Barre should experience no change with regards to land value taxation. Observed Over the past 20 years, rents in Altoona, Wilkes Barre, and Pittsburgh have all moved in the same direction in terms of gross rents in renter occupied housing units rising as shown in Table 4 5 Gross Rents Renting in Pittsburgh has been most expensive among the three study areas, with Altoona maintaining a slightly more affordable renters market than Wilkes Barre. Between 2000 and 2009, Pittsburgh also saw a sharp increase of 13% in $1000/month gross rent units, compared to just below 4% in Altoona and Wilkes Barre. Determination Gross rents have not produced evidence of the theoretical effects of land value taxation. While Pittsburgh experienced gross rent in creases as predicted by theory, is complex and while the tax policy may have had in impact on gross rents it is not overtly clear in the U S Census and ASC data. Ag rents increased in the 2001 2009 period. As a result of the abatements effectively replacing the land value tax, we would not expect gross rents in Pittsburgh to rebound like they did. It is pos revitalization centered on higher wage employment like health care and education have caused increases in the most expensive categories of gross rent. While failing to explain the entire city, this eme rgence of high wage employment, particularly downtown, may be responsible for increases in the amount of high end units.

PAGE 76

76 employment like health care and education have caused increases in the most expensive categories of gross rent. While failing to explain the entire city, this emergence of high wage employment, particularly downtown, may be responsible for increases in the amount of high end units. increased gross rents could potentially be explained by growth in higher wage industries as well, despite more modest gains in ve ry high wage employment in the c ity compared to Pittsburgh. Wilkes in a similar way to Altoona duri ng the 2001 2009 period and both markets experienced decline in the once dominant manufacturing industry and growth in the health care and education industries. Variable Year Built Expectations b ased on t heory ousing stock is unclear. It is clear that the adoption of land value taxation should lead to an increase in improvements per area. However, it is unclear how that increase is to occur. Assuming the housing stock is ill equipped to handle the increase in improvements per acre as additions to already existing structures, the adoption of land value taxation is likely to result in the demolition of structures of few units in favor of multi unit new construction. This effect would make housing stock younger. However, if housing stock can absorb the increase in improvements per acre as additions to pre existing structures, the adoption of the policy is likely to prolong the life of such buildings and increases the age of structures. As a result, should curren t housing stock be unsuitable to the addition of new units, Altoona should expect to see a younger housing stock by 2009. Pittsburgh

PAGE 77

77 should expect to see an older housing stock and Wilkes Barre should expect no change in the age of its housing stock as a result of land value taxation. Observed By 2009, data from Table 4 6 Year Structure Built indicates that housing stock in Altoona became slightly younger than in Pittsburgh and Wilkes Barre. Pre 1940 housing units in Altoona decreased over the period whi le such units increased in Pittsburgh and Wilkes Barre. There was slightly more new construction in Pittsburgh than Altoona and nearly zero new construction in Wilkes Barre. Determination Pittsburgh, housing stock has been realized. As stated in the theoretical analysis of the variable, ge of housing stock. The varied and numerous characteristics of housing stock and local housing markets are of too large of scope to be covered in this paper. The fact that both Pittsburgh and Wilkes Barre experienced percentage increases in their pre 19 40s housing may be an indication that such housing was well built, perhaps constructed better than in Altoona. housing stock relative to Pittsburgh and Wilkes Barre. Pittsburgh saw a boom in housing unit construction during the 1950s and Wilkes Barre saw the same in the 1970s. These booms are still evident in 2009 ACS data with a higher percentage of 1950s housing than from any other decade since 1940 in Pittsburgh and a greater share of housing from the 1970s than any other decade since 1940. A pre 1940s housing

PAGE 78

78 boom may have occurred in Altoona that may have been removed from the market between 2001 and 2009. in its economy grew faster in terms of employment than the economies of Pittsburgh and Wilkes Barre. Older housing is often suited to fit the needs of lower income household s by way of filtering. Despite new (2000 2009) housing construction being greatest in Pittsburgh, older housing in Pittsburgh and Wilkes Barre may have remained in the market while similar housing in Altoona dropped out. Variable Density Measures Expecta tions based on t heory One of the central tenets of land value taxation is its ability to increase density. This theory is supported by Oates & Schwab (2009) and Brueckner & Kim (2003). With the relative tax rate reduced on improvements and increased on l and, more improvements will be consumed and less land will be consumed. Thus, density will increase. value taxation will lead to increased employment per area ratios and, in t urn, will cause more workers to move to the area faster than building high can increase. As a result, lan d value taxation will lead to in creases in density in the short term and enlargement of evelopment Pattern Should the former theory hold true, we would expect Altoona to have increased density and Pittsburgh to have decreased density by 2009. Density in Wilkes Barre

PAGE 79

79 would have not been expected to change. If the latter theory should hold true, Altoona would expect to experience a decrease in density by 2009 with Pittsburgh experiencing an increase in density. Again, Wilkes Barre would not be expected to see a change in density as a result of land value taxation. For the sake of this experiment one of these theories will be chosen as the hypothesis. It will be assumed that the reduced tax rate on improvements relative to the tax rate on land will encourage more improvements to be consumed and less land. Thus, dens ity will increase. Observed According to Table 4 7 Density Measures, between 2000 and 2009, all three cities densities fell by 1.54% and 6.01%, respectively. It was the only area to experience a substantial drop in unit density. Overall, Altoona saw its density fall the most over the 20 year period. Pittsburgh appeared to recapture unit density growth during the 2000s. All three study areas continue to experience populatio n density decline, however that rate of density decline between in 2000 and 2009 increased only in Altoona. Determination value taxation while Pittsburgh actually experienced a unit density increase over the 8 years following its termination. Therefore, density measures of the three case study areas do not offer evidence that the adoption of land value taxation increases density Additionally, even if the Mills (1998) theory th at land value taxation increases density in the short term and decreases it in the long term is assumed true, it would not be clear that it has manifested itself in this data. Altoona had only put in place land

PAGE 80

80 value taxation policy for seven years by 200 9 and differentiation from short and long term would be difficult to discern. P opulation and unit density are some of the variables most directly related to the property tax sch eme. T his deviation from theoretical ly expected effects makes it even more di fficult to argue that the presence of land value taxation has significantly impacted the U S Census and ASC data of Altoona or of Pittsburgh. For an explanation of why theoretical effects on density were not presented in Pittsburgh, the tax abatement poli cies may again offer solutions. As previously property tax abatement policies have filled the void of the tax regime by effectively reducing the tax rate on improv ements relative to the tax rate on land. As a result, the abatements encourage more improvements to be consumed and less land, perpetuating increased density. could be the result of economy local economy has replaced generally lower wage employment with higher wage employment. As income increases, demand for larger properties and more privacy tend to increase as well. As a result, the change in demographics away from low wage earners may have impacted the decline in density. Income Variables This section will compare theoretically expectations with observed results for income variables. There are a total of three inco me variables in this study.

PAGE 81

81 Variable Household Income Expectations based on t heory The theoretical relationship between household income and land value taxation is not overtly clear. Land value taxation is directly related to improvement (i.e. housing, etc.) and land markets. However, reduction in deadweight loss in economic systems is an impact that land value taxation makes that can be correlated to income. This reduction in deadweight loss improves the efficiency of economic systems, increasing weal th generation and likely median income. For more information on deadweight loss, Accordingly, in 2000, Pittsburgh would be expected to have the highest median household income. At that time, Pittsburgh had implemented land value taxation while Altoona and Wilkes have the most efficient local economy. By 2009, higher median income would be expected to grow in Altoona than in Pittsburgh. At t hat time, Altoona had adopted land value taxation while Pittsburgh had rescinded it. Altoona should also have a higher median income than Wilkes Barre based on property tax scheme alone. The increases in ma rket efficiency should result from less deadweig ht loss. Therefore, higher profits and higher incomes should be expected in areas using land value taxation. Observed According to 2000 data from Table 4 8 Household Income, all three case study areas experienced roughly the same rate of median household income growth during the 1990s. Pittsburgh experienced more growth in households with incomes of

PAGE 82

82 $100,000 or more than Altoona or Wilkes Barre and had the highest median household income at $28,588. Table 4 8 Household Income also shows that the median h ousehold income in Pittsburgh and Altoona grew from 2000 to 2009, but at slower rates than during the only 19.03%. Wilkes same as it grew by a mere 7.44% between 2000 and 2009. Dete rmination Data from Table 4 8 Household Income matches the theoretical expectation in the year 2000 as Pittsburgh has the highest median income of the three study areas and also grew the most during the preceding decade. However, 2009 data from Table 4 9 Median Household Income is not congruent with theory as Altoona has actually fallen further behind Pittsburgh in terms of median household income and median household income growth. Altoona did outp erform Wilkes Barre in this regard over the period, however, which theory would likely predict. Essentially acting like land value taxation, the tax abatements also remo ve deadweight loss from the economic system and increase economic efficiency. Pittsburgh failed to relinquish its ability to reduce deadweight loss and therefore, its maintenance of relatively high median household income is unsurprising. In Altoona, stro ng increases in median household income were likely tempered by its relatively small and less diverse economy in comparison with Pittsburgh. Both cities income increased more t

PAGE 83

83 in its share of national employment while Pittsburgh did not. Therefore, it may be concluded that while Altoona experienced rises in its share of national employment, employment in high wage sect ors, and median household income, it did not attract the Over the study period, Altoona also maintained a residual p resence of low wage and low growth industries despite increased significance in the health care and administrative support and waste services industries. According to shift share analysis, reliant upon the manufa cturing and construction sectors that experience d slow growth over the period. While a shift to from blue to white collar work continued to occur in Altoona, the remaining presence of low wage employment may have dampened median income growth. Variable Per Capita Income Expectations b ase d on t heory The same theoretical argument associated with the impact of land value taxation for household income holds true for per capita income. Assuming that reducing deadweight loss improves the efficiency of economi c systems and therefore increases wealth and income generation, municipalities with land value taxation in place should experience higher per capita income than those municipalities without it. As in household income, in 2000, Pittsburgh would be expected to have the highest median per capita income as it was the only one of the case studies implementing land value taxation at that time. By 2009, higher per capita income would be expected in Altoona than in Pittsburgh. Altoona should also have a higher per capita income than Wilkes Barre

PAGE 84

84 based on property tax scheme alone. The increases in market efficiency should result in higher profits and higher incomes in areas using land value taxation. Observed Acc ording to 2000 data from Table 4 10 Per Capita Inco me, Pittsburgh had the highest overall per capita income and highest per capita income growth rate, followed in both respects by Altoona and then Wilkes Barre. By 2009, Table 4 10 shows that Pittsburgh continued to have the highest per capita income and hi ghest per capita income growth rate from 2000 to 2009 of the three case study areas. Again this was followed by Altoona then Wilkes Barre. Also of note, all three study areas experienced decreased per capita income growth rates. Determination Just as in the examination of household income data, observations in the year 2000 matched expectations but failed to do so outcomes for the two municipalities. Also just as in the examination of household income data, the impact of reduction should spell continued high income levels. The observations in Altoon a could also be explained by the lingering presence of relatively low wage employment and lack of very high wage employment. These factors may have arisen from the relatively small size and industrial homogeneity of the market in Altoona.

PAGE 85

85 Variable Incom e to Poverty Ratio Expectations based on t heory Income to poverty ratio is calculated by dividing a family's or person's income by their poverty threshold Values for the ratio of income to poverty level measure an tios below 1.0 indicate poverty, with ratios .5 indicating 11 Ratio of Income to Poverty Level shows the amount of individuals within particul ar ratio ranges from the three case study areas. The same income arguments pertain to this theoretical expectation. As previously stated, economic efficiency gains should be the tide that raise s all incomes on one side of this ratio. A more efficient eco nomy should lead to more efficient wealth distribution. Progress and Poverty had as much to do with economic efficiency as with wealth distribution and poverty reduction. Therefore, in 2000, we would expec t that Pittsburgh would have the highest income to poverty ratios. Altoona should be expected to experience improvements in this ratio by 2009 after policy intervention. Pittsburgh should theoretically drop off in this ratio as they reduce economic effic iency and reduce the ability of low income residents to compete for housing based on size. The income to poverty ratio in Wilkes Barre should not be affected by land value taxation. Observed Pittsburgh had the least healthy income to poverty ratio figur es during the 1990s according to Table 4 11 Income to Poverty Ratio. The city had the highest rates indicating poverty (ratios below 1) and severe poverty (ratios below .5).

PAGE 86

86 By 2009, all three study areas showed worsening income to poverty ratios. Howeve r, Altoona faired the best of the three over the first decade of the 21 st century where ratios indicating severe poverty fell by 1.98%. Pittsburgh saw small increases in ratios indicating poverty and severe poverty but figures were worst in Wilkes Barre w here ratios indicating poverty increased by 6% and severe poverty by over 4%. Determination Data for this variable from 2000 does not match theory as Pittsburgh, which should have performed best, performed worst. Income to poverty ratios in Altoona and Wilke s Barre were similar which was anticipated by theory. Data from 2000 to 2009 correlated better with theory. While income to poverty ratios declined across the board, Altoona performed better than Pittsburgh and far better than Wilkes Barre. The fact that income to poverty ratios performed better in Altoona and Pittsburgh than in Wilkes Barre between 2000 and 2009 is an encouraging sign for the prospects of land value taxation. This data may show that the reduction in deadweight loss, emanating from Altoon shift from the regime to tax abatement policies, allowed the two cities to outperform Wilkes Barre. Summary In this chapter, we have examined expected theoretical results of land value taxatio n and their observed counterparts from Altoona, Pittsburgh and Wilkes Barre, PA. Data for selected housing and income characteristics from the 1990 and 2000 U S Census and the 2005 2009 ACS was compiled to examine land value taxation as a policy interven tion on affecting important variables. Comparing the theoretical effects

PAGE 87

87 with cross sectional observations conclusions a n d d e t e r m i n a t i o n s have been drawn as to whether the impact of land value taxation has been made evident in U S Census and ACS data. The next and final c hapter, Chapter 5 Findings and Conclusions, will discuss the results holistically, final conclusions will be drawn interpreting the effects of land value taxation on U S Census and ACS data in the three case study areas.

PAGE 88

88 Table 4 1 Total housing u nits 1990 2000 Change % 2009 Change % Altoona city, Pennsylvania 22,698 21,682 4.48 21,348 1.54 Pittsburgh city, Pennsylvania 170,159 163,366 3.99 165,294 1.18 Wilk es Barre city, Pennsylvania 20,734 20,294 2.12 20,245 0.24 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 89

89 Table 4 2 Occupancy s tatus Altoona city, Pe nnsylvania 1990 2000 Change % 2009 Change % Occupied 20,684 20,060 1.39 19,316 2.04 Vacant 2,014 1,622 1.39 2,032 2.04 Total 22,698 21,682 4.48 21,348 1.54 Pittsburgh city, Pennsylvania 1990 2000 Change % 2009 Change % Occupied 153,483 143, 739 2.21 138,739 4.05 Vacant 16,676 19,627 2.21 26,555 4.05 Total 170,159 163,366 3.99 165,294 1.18 Wilkes Barre city, Pennsylvania 1990 2000 Change % 2009 Change % Occupied 19,435 17,961 5.23 17,442 2.35 Vacant 1,299 2,333 5.23 2,803 2.35 Total 20,734 20,294 2.12 20,245 0.24 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 90

90 Table 4 3 Units in s tructure Altoona city, Pennsylvania 1990 2 000 Change % 2009 Change % 1, detached 15,818 15,028 0.38 15,009 1.00 1, attached 1,141 941 0.69 938 0.05 2 2,104 1,871 0.64 1,826 0.08 3 or 4 1,131 1,219 0.64 1,127 0.34 5 to 9 798 866 0.48 684 0.79 10 to 19 553 469 0.27 653 0.90 20 to 49 30 2 265 0.11 180 0.38 50 or more 549 996 2.17 870 0.52 Mobile home or trailer 35 27 0.03 61 0.16 Other 267 0 1.18 0 0.00 Total 22,698 21,682 4.48 21,348 1.54 Pittsburgh city, Pennsylvania 1990 2000 Change % 2009 Change % 1, detached 70,0 48 71,570 2.64 73,751 0.81 1, attached 27,756 24,277 1.45 26,736 1.31 2 15,965 15,894 0.35 14,512 0.95 3 or 4 13,302 12,749 0.01 12,369 0.32 5 to 9 12,235 10,818 0.57 10,723 0.13 10 to 19 8,496 7,794 0.22 6,992 0.54 20 to 49 6,616 5,499 0.52 6,347 0.47 50 or more 13,725 14,382 0.74 13,186 0.83 Mobile home or trailer 422 354 0.03 668 0.19 Other 1,594 29 0.92 10 0.01 Total 170,159 163,366 3.99 165 294 1.18

PAGE 91

91 Table 4 3 Continued Wilkes Barre city, Pennsylvania 1990 2000 Change % 200 9 Change % 1, detached 9,018 9,395 2.80 9,515 0.70 1, attached 3,147 2,573 2.50 3,507 4.64 2 2,111 2,295 1.13 2,257 0.16 3 or 4 1,860 2,008 0.92 1,395 3.00 5 to 9 1,208 1,175 0.04 906 1.31 10 to 19 495 454 0.15 501 0.24 20 to 49 883 693 0.84 446 1.21 50 or more 1,612 1,689 0.55 1,672 0.06 Mobile home or trailer 14 12 0.01 46 0.17 Other 386 0 1.86 0 0.00 Total 20,734 20,294 2.12 20,245 0.24 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Populat ion and Housing 2005 2009 American Community Survey

PAGE 92

92 Table 4 4 Value of specified owner occupied housing u nits Altoona city, Pennsylvania 1990 2000 Change % 2009 Change % Less than $15,000 1,760 358 10.78 231 1.02 $15,000 to $19,999 1,597 306 9. 93 138 1.34 $20,000 to $24,999 1,505 421 8.30 278 1.15 $25,000 to $29,999 1,440 696 5.61 199 3.97 $30,000 to $34,999 1,257 751 3.75 399 2.82 $35,000 to $39,999 919 878 0.12 430 3.59 $40,000 to $49,999 1,624 1,604 0.21 1,050 4.45 $50,000 to $59,999 1,068 1,572 4.27 1,316 2.09 $60,000 to $69,999 1,563 1,304 2.11 $60,000 to $74,999 945 2157.5 9.89 1973 1.54 $70,000 to $79,999 1,189 1,338 1.14 $80,000 to $89,999 946 1,330 3.02 $90,000 to $99,999 720 1,126 3.20 $75,000 to $99,999 504 2260.5 14.13 3125 6.79 $100,000 to $124,999 128 795 5.35 1,233 3.45 $125,000 to $149,999 69 385 2.54 978 4.70 $150,000 to $174,999 80 122 0.35 540 3.32 $175,000 to $199,999 0 100 0.80 286 1.47 $200,000 to $249,999 0 68 0.54 150 0.65 $250,000 to $299,999 9 21 0.10 115 0.75 $300,000 to $399,999 0 33 0.26 112 0.63 $400,000 to $499,999 0 0 0.00 15 0.12 $500,000 or more 0 7 0.06 20 0.10 Total: 12,905 12,535 2.87 12,588 0.42

PAGE 93

93 Table 4 4 Continued Pittsburgh city, Pennsylvania 1990 2000 Change % 2009 Change % Less than $15,000 5,033 1,909 4.31 1155 1.27 $15,000 to $19,999 4,786 1,894 3.98 886 1.62 $20,000 to $24,999 5,008 2,260 3.75 1,437 1.41 $25,000 to $29,999 5,920 2,876 4.13 2,105 1.41 $30,000 to $34,999 6,715 3,741 3.96 1,829 3.09 $35,000 to $39,999 6,970 4,542 3.12 2,487 3.38 $40,000 to $49,999 11,817 8,253 4.46 4,941 5.56 $50,000 to $59,999 7,924 8,050 0.79 6,037 3.74 $60,000 to $69,999 7,693 6,570 2.47 $60,000 to $74,999 6,600 10839.5 6.87 9924.5 2.55 $70,00 0 to $79,999 6,293 6,709 0.17 $80,000 to $89,999 4,684 6,436 1.87 $90,000 to $99,999 2,989 4,243 1.38 $75,000 to $99,999 3,962 10819.5 10.60 14033.5 3.16 $100,000 to $124,999 1,187 3,495 3.56 6,813 4.17 $125,000 to $149,999 941 2,060 1.75 4,13 6 2.63 $150,000 to $174,999 604 1,047 0.71 3,663 3.49 $175,000 to $199,999 596 976 0.62 2,018 1.33 $200,000 to $249,999 674 1,012 0.56 2,957 2.57 $250,000 to $299,999 406 803 0.63 1,906 1.43 $300,000 to $399,999 398 840 0.69 2,771 2.57 $400,000 to $4 99,999 222 450 0.36 1,275 1.09 $500,000 or more 320 701 0.60 1,912 1.59 Total: 70,083 66,568 5.02 72,286 8.59

PAGE 94

94 Table 4 4 Continued Wilkes Barre city, Pennsylvania 1990 2000 Change % 2009 Change % Less than $15,000 253 62 2.12 78 0.14 $15,000 to $19,999 381 83 3.31 16 0.78 $20,000 to $24,999 433 124 3.42 99 0.34 $25,000 to $29,999 729 163 6.29 95 0.83 $30,000 to $34,999 903 192 7.90 281 0.88 $35,000 to $39,999 934 402 5.83 284 1.50 $40,000 to $49,999 1,970 1,149 8.83 553 7.14 $50, 000 to $59,999 1,195 1,461 3.43 933 6.56 $60,000 to $69,999 1,495 1,661 1.05 $60,000 to $74,999 1,086 2,039 11.30 2,240 1.16 $70,000 to $79,999 1,087 1,158 0.22 $80,000 to $89,999 854 922 0.31 $90,000 to $99,999 617 831 2.03 $75,000 to $99,9 99 665 2,015 15.73 2,332 2.45 $100,000 to $124,999 165 496 3.86 967 4.92 $125,000 to $149,999 105 218 1.33 451 2.45 $150,000 to $174,999 26 106 0.93 360 2.74 $175,000 to $199,999 38 55 0.21 132 0.82 $200,000 to $249,999 37 92 0.64 97 0.01 $250,000 t o $299,999 12 26 0.16 113 0.94 $300,000 to $399,999 0 4 0.05 43 0.43 $400,000 to $499,999 0 5 0.06 11 0.06 $500,000 or more 0 0 0.00 14 0.15 Total: 8,932 8,691 2.70 9,099 4.69

PAGE 95

95 Table 4 5 Gross r ent Altoona city, Pennsylvania With cash rent: 199 0 2000 Change % 2009 Change % Less than $100 191 200 0.20 58 2.06 $100 to $149 582 305 3.85 69 3.44 $150 to $199 639 511 1.65 478 0.37 $200 to $249 956 417 7.56 353 0.85 $250 to $299 1,382 555 11.63 326 3.27 $300 to $349 993 693 4.05 256 6 .33 $350 to $399 605 900 4.52 426 6.84 $400 to $449 564 738 2.74 576 2.24 $450 to $499 337 558 3.35 789 3.56 $500 to $549 176 496 4.74 612 1.84 $550 to $599 95 305 3.10 280 0.30 $600 to $649 48 322 4.03 449 1.96 $650 to $699 28 186 2.32 323 2.08 $700 to $749 7 87 1.17 240 2.29 $750 to $999 15 115 1.47 856 11.04 $1,000 or more 42 94 0.78 346 3.77 No cash rent 339 353 0.32 291 0.84 Total: 6,999 6,835 6728

PAGE 96

96 Table 4 5 Continued Pittsburgh city, Pennsylvania With cash rent: 1990 2000 Chan ge % 2009 Change % Less than $100 3,223 1,764 1.85 524 1.78 $100 to $149 5,645 2,346 4.32 1,002 1.91 $150 to $199 4,068 3,155 0.98 1,889 1.75 $200 to $249 4,340 2,527 2.27 2,047 0.60 $250 to $299 6,145 2,359 4.99 1,732 0.83 $300 to $349 8,4 71 3,760 6.14 1,558 3.13 $350 to $399 9,598 4,470 6.65 1,469 4.30 $400 to $449 7,671 6,250 1.41 2,414 5.47 $450 to $499 6,106 6,440 1.01 3,052 4.78 $500 to $549 4,640 6,294 2.80 3,906 3.28 $550 to $599 2,999 5,208 3.47 4,085 1.43 $600 to $64 9 2,389 4,649 3.49 4,833 0.50 $650 to $699 1,546 3,753 3.34 4,315 1.03 $700 to $749 997 2,866 2.81 3,940 1.76 $750 to $999 2,266 7,063 7.18 15,017 12.32 $1,000 or more 662 3,141 3.67 11,690 13.02 No cash rent 2,203 2,647 0.83 2,980 0.63 Total: 72,969 68 692 66 453

PAGE 97

97 Table 4 5 Continued Wilkes Barre city, Pennsylvania With cash rent: 1990 2000 Change % 2009 Change % Less than $100 302 208 0.80 210 0.02 $100 to $149 744 307 4.42 75 2.78 $150 to $199 789 521 2.35 355 1.99 $200 to $249 923 444 4.73 303 1.69 $250 to $299 1,201 461 7.56 355 1.27 $300 to $349 1,273 826 3.97 266 6.71 $350 to $399 1,407 1,003 3.30 266 8.83 $400 to $449 719 913 3.11 506 4.88 $450 to $499 478 860 5.10 524 4.03 $500 to $549 379 536 2.30 663 1.52 $5 50 to $599 230 555 4.15 726 2.05 $600 to $649 124 332 2.63 667 4.02 $650 to $699 63 327 3.23 457 1.56 $700 to $749 50 214 2.02 666 5.42 $750 to $999 70 297 2.80 1,254 11.47 $1,000 or more 37 137 1.24 460 3.87 No cash rent 392 403 0.56 590 2.24 Total : 9,181 8,344 8343 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 98

98 Table 4 6 Year structure b uilt Altoona city, Pennsylvania 1990 % 2000 % % Change 2009 % % Change 2005 or later 0 0.00 0 0.00 0.00 25 0.12 0.12 2000 to 2004 0 0.00 0 0.00 0.00 296 1.39 1.39 1990 to 1999 0 0.00 729 3.36 3.36 753 3.53 0.17 1980 to 1989 827 3.64 642 2.96 0.68 663 3.11 0.14 1970 to 1979 2,108 9.29 1,779 8.20 1.08 1, 721 8.06 0.14 1960 to 1969 1,647 7.26 1,508 6.96 0.30 1,765 8.27 1.31 1950 to 1959 1,796 7.91 2,249 10.37 2.46 2,375 11.13 0.75 1940 to 1949 2,254 9.93 2,702 12.46 2.53 2,581 12.09 0.37 1939 or earlier 14,066 61.97 12,073 55.68 6.29 11,169 52.32 3 .36 Total 22,698 100 21,682 100 4.48 21,348 100.00 1.54

PAGE 99

99 Table 4 6. Continued Pittsburgh city, Pennsylvania 1990 % 2000 % % Change 2009 % % Change 2005 or later 0 0.00 0 0.00 0.00 1,181 0.71 0.71 2000 to 2004 0 0.00 0 0.00 0.00 2,994 1.81 1.81 19 90 to 1999 0 0.00 3,834 2.35 2.35 4,227 2.56 0.21 1980 to 1989 8,015 4.71 5,925 3.63 1.08 6,753 4.09 0.46 1970 to 1979 9,919 5.83 10,275 6.29 0.46 10,949 6.62 0.33 1960 to 1969 15,627 9.18 15,513 9.50 0.31 12,046 7.29 2.21 1950 to 1959 21,276 12.50 2 2,896 14.02 1.51 22,437 13.57 0.44 1940 to 1949 21,269 12.50 22,152 13.56 1.06 17,066 10.32 3.24 1939 or earlier 94,053 55.27 82,771 50.67 4.61 87,641 53.02 2.36 Total 170,159 100 163,366 100 3.99 165,294 100 1.18

PAGE 100

100 Table 4 6 Continued Wilkes Ba rre city, Pennsylvania 1990 % 2000 % % Change 2009 % % Change 2005 or later 0 0.00 0 0.00 0.00 20 0.10 0.10 2000 to 2004 0 0.00 0 0.00 0.00 61 0.30 0.30 1990 to 1999 0 0.00 252 1.24 1.24 276 1.36 0.12 1980 to 1989 391 1.89 515 2.54 0.65 760 3.75 1.22 1970 to 1979 3,408 16.44 2,877 14.18 2.26 2,089 10.32 3.86 1960 to 1969 1,098 5.30 1,292 6.37 1.07 1,102 5.44 0.92 1950 to 1959 1,001 4.83 1,136 5.60 0.77 1,321 6.53 0.93 1940 to 1949 1,515 7.31 1,993 9.82 2.51 1,561 7.71 2.11 1939 or earlier 13,32 1 64.25 12,229 60.26 3.99 13,055 64.49 4.23 Total 20,734 100 20,294 100 2.12 20,245 100 0.24 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 101

101 Table 4 7. Density m easurements Altoona city, Pennsylvania Year 1990 2000 % Change 2009 % Change Unit Density 2,323.62 2,219.61 4.48 2,185.42 1.54 Population Density 5,311.12 5,069.93 4.54 4,765.48 6.01 Pittsburgh city, Pennsylvania Year 1990 2000 % Change 2009 % Change Unit Density 3,061.27 2,939.06 3.99 2,973.75 1.18 Population Density 6,654.36 6,019.01 9.55 5,633.20 6.41 Wilkes Barre city, Pennsylvania Year 1990 2000 % Change 2009 % Change Unit Density 3 ,027.33 2,963.09 2.12 2,955.94 0.24 Population Density 6,938.75 6,296.31 9.26 5,998.60 4.73 US Year 1990 2000 % Change 2009 % Change Unit Density 70.31 79.56 13.15 85.22 7.12 Population Density 28.91 32.77 13.34 36.1 0 10.18 Source: U. S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 102

102 Table 4 8 Household i ncome Altoona city, Pennsylvania 1990 2000 Change % 2009 Change % Less than $5,000 1,646 $5,000 to $9,999 3,311 Less than $10,000 4,957 2,990 9.15 2,250 3.23 $10,000 to $12,499 1,568 $12,500 to $14,999 1,178 $10,000 to $14,999 2,746 2,280 1.97 1,781 2.13 $15,000 to $17,499 1,241 $17,500 to $19,9 99 1,046 $15,000 to $19,999 2,287 1,920 1.53 1,825 0.11 $20,000 to $22,499 1,157 $22,500 to $24,999 924 $20,000 to $24,999 2,081 1,723 1.51 1,285 1.92 $25,000 to $27,499 863 $27,500 to $29,999 957 $25,000 to $29,999 1,820 1,6 02 0.85 1,563 0.12 $30,000 to $32,499 1,024 $32,500 to $34,999 762 $30,000 to $34,999 1,786 1,342 1.98 1,356 0.34 $35,000 to $37,499 711 $37,500 to $39,999 491 $35,000 to $39,999 1,202 1,362 0.95 927 1.98 $40,000 to $42,499 520 $42,500 to $44,999 463 $40,000 to $44,999 983 1,207 1.24 815 1.79 $45,000 to $47,499 435 $47,500 to $49,999 312 $45,000 to $49,999 747 926 0.99 1,048 0.82 $50,000 to $54,999 496

PAGE 103

103 Table 4 8. Continued Altoona city, Pennsylvania 1990 2000 Change % 2009 Change % $55,000 to $59,999 342 $50,000 to $59,999 838 1,608 3.94 1,637 0.47 $60,000 to $74,999 687 1,375 3.51 1,845 2.71 $75,000 to $99,999 252 1,071 4.11 1,785 3.91 $100,000 to $124,999 104 390 1.44 706 1.71 $125,0 00 to $149,999 27 119 0.46 322 1.07 $150,000 or more 106 176 0.36 171 0.01 Total: 20,623 20,091 2.58 19,316 3.86

PAGE 104

104 Table 4 8 Continued Pittsburgh city, Pennsylvania 1990 2000 Change % 2009 Change % Less than $5,000 16,714 $5,000 to $9,99 9 24,399 Less than $10,000 41,113 25,927 8.73 19,523 3.96 $10,000 to $12,499 9,858 $12,500 to $14,999 7,825 $10,000 to $14,999 17,683 13,668 2.00 11,561 1.18 $15,000 to $17,499 8,367 $17,500 to $19,999 7,365 $15,000 to $19,99 9 15,732 12,657 1.44 10,372 1.33 $20,000 to $22,499 7,619 $22,500 to $24,999 6,143 $20,000 to $24,999 13,762 11,949 0.65 10,008 1.10 $25,000 to $27,499 6,654 $27,500 to $29,999 5,604 $25,000 to $29,999 12,258 10,074 0.97 8,711 0 .73 $30,000 to $32,499 5,995 $32,500 to $34,999 4,471 $30,000 to $34,999 10,466 9,154 0.45 8,123 0.51 $35,000 to $37,499 4,846 $37,500 to $39,999 3,662 $35,000 to $39,999 8,508 8,065 0.07 6,998 0.57 $40,000 to $42,499 3,804 $ 42,500 to $44,999 2,935 $40,000 to $44,999 6,739 7,163 0.60 6,863 0.04 $45,000 to $47,499 3,133 $47,500 to $49,999 2,020 $45,000 to $49,999 5,153 6,213 0.97 6,266 0.19 $50,000 to $54,999 4,218

PAGE 105

105 Table 4 8. Continued Pittsburgh city, Pennsylvania 1990 2000 Change % 2009 Change % $55,000 to $59,999 3,214 $50,000 to $59,999 7,432 9,788 1.97 10,126 0.49 $60,000 to $74,999 6,285 10,694 3.35 11,404 0.78 $75,000 to $99,999 3,972 8,366 3.23 12,176 2.96 $100,000 to $124,999 1,673 4,084 1.75 6,472 1.82 $125,000 to $149,999 805 1,759 0.70 3,235 1.11 $150,000 or more 2,026 4,191 1.60 6,901 2.06 Total: 153,607 143,752 6.42 138,739 3.49

PAGE 106

106 Table 4 8 Continued Wilkes Barre city, Pennsylvania 1990 2000 Change % 2009 Chan ge % Less than $5,000 1,612 $5,000 to $9,999 3,367 Less than $10,000 4,979 3,070 8.69 2,786 1.16 $10,000 to $12,499 1,500 $12,500 to $14,999 1,163 $10,000 to $14,999 2,663 2,150 1.81 1,908 1.06 $15,000 to $17,499 1,228 $17,5 00 to $19,999 953 $15,000 to $19,999 2,181 1,803 1.25 1,559 1.12 $20,000 to $22,499 976 $22,500 to $24,999 753 $20,000 to $24,999 1,729 1,458 0.83 1,480 0.35 $25,000 to $27,499 968 $27,500 to $29,999 700 $25,000 to $29,999 1,6 68 1,446 0.58 1,308 0.57 $30,000 to $32,499 866 $32,500 to $34,999 543 $30,000 to $34,999 1,409 1,216 0.52 1,113 0.40 $35,000 to $37,499 673 $37,500 to $39,999 505 $35,000 to $39,999 1,178 1,090 0.03 766 1.69 $40,000 to $42,499 548 $42,500 to $44,999 398 $40,000 to $44,999 946 968 0.50 715 1.30 $45,000 to $47,499 343 $47,500 to $49,999 359 $45,000 to $49,999 702 796 0.80 470 1.75 $50,000 to $54,999 455

PAGE 107

107 Table 4 8. Continued Wilkes Barre city, Pennsylvania 1990 2000 Change % 2009 Change % $55,000 to $59,999 334 $50,000 to $59,999 789 1,165 2.41 1,524 2.24 $60,000 to $74,999 550 1,156 3.60 1,320 1.12 $75,000 to $99,999 290 1,011 4.14 1,487 2.88 $100,000 to $124,999 119 324 1.19 520 1. 17 $125,000 to $149,999 7 95 0.49 325 1.33 $150,000 or more 74 171 0.57 161 0.03 Total: 19,284 17919 7.08 17442 2.66 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 108

108 Table 4 9 Median household i ncome 1990 2000 Change % 2009 Change % Altoona city, Pennsylvania 20,695 28,248 36.50 33,623 19.03 Pittsburgh city, Pennsylvania 20,747 28,588 37.79 35,732 24.99 Wilkes Barre city, Pennsylvania 19,52 5 26,711 36.80 28,699 7.44 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 109

109 Table 4 10 Per capita i ncome 1990 2000 Change % 2009 Change % Altoona city, Pennsylvania 10 398 15,213 46.31 18,272 20.11 Pittsburgh city, Pennsylvania 12 580 18,816 49.57 24,616 30.82 Wilkes Barre city, Pennsylvania 10 513 15,050 43.16 17,171 14.09 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2 000 Census of Population and Housing 2005 2009 American Community Survey

PAGE 110

110 Table 4 11 Ratio of income to poverty l evel Altoona city, Pennsylvania 1990 % 2000 % Change % 2009 % Change % Under .50 3,899 7.62 3,522 7.34 0.29 2,386 5.36 1.98 .5 0 to .74 2,532 4.95 2,340 4.87 0.00 .75 to .99 2,787 5.45 2,634 5.49 0.00 .50 to .99 5,319 10.40 4,974 10.36 0.04 6,050 13.58 3.22 Under 1.00 9,218 18.03 8,496 17.69 0.33 8,436 18.94 1.25 1.00 to 1.24 2,695 5.27 2,917 6.08 0.81 3,135 7.04 0.96 1.25 to 1.49 3,343 6.54 3,157 6.58 0.04 2,636 5.92 0.66 1.50 to 1.74 2,888 5.65 2,633 5.48 0.00 1.75 to 1.84 1,226 2.40 1,223 2.55 0.00 1.50 to 1.84 4,114 8.04 3,856 8.03 0.01 3,681 8.26 0.23 1.85 to 1.99 1,932 3.78 1,533 3.19 0.59 1,034 2.32 0.87 2.00 and over 29,837 58.34 28,056 58.43 0.09 25,619 57.52 0.91 1.00 and over 41,921 81.97 39,519 82.31 0.33 36,105 81.06 1.25 Total Pop. 51,139 48,015 6.11 44,541 7.24

PAGE 111

111 Table 4 11. Continued Pittsburgh city, Pennsylvania 1990 % 2000 % Change % 2009 % Change % Under .50 39,589 11.28 32,551 10.39 0.89 30,889 10.56 0.18 .50 to .74 17,006 4.85 15,865 5.06 0.00 .75 to .99 18,577 5.29 15,450 4.93 0.00 .50 to .99 35,583 10.14 31,315 9.99 0.15 32,453 11.10 1.11 Under 1.00 75,172 21.42 63,866 20.38 1.04 63,342 21.66 1.28 1.00 to 1.24 18,438 5.25 16,121 5.14 0.11 14,791 5.06 0.09 1.25 to 1.49 17,064 4.86 16,542 5.28 0.42 14,628 5.00 0.28 1.50 to 1.74 17,895 5.10 15,848 5.06 0.00 1.75 to 1.84 7,021 2.00 6,734 2.15 0.00 1.50 to 1.84 24,916 7.10 22,582 7.21 0.11 20,139 6.89 0.32 1.85 to 1.99 10,971 3.13 9,525 3.04 0.09 9,409 3.22 0.18 2.00 and over 204,379 58.24 184,747 58.95 0.71 170,090 58.17 0.78 1.00 and over 275,768 78.58 249,517 79.62 1.04 229,057 78.34 1.28 Total Pop. 350,940 313,383 10.70 292,399 6.70

PAGE 112

112 Table 4 11 Continued Wilkes Barre city, Pennsylvania 1990 % 2000 % Change % 2009 % Change % Under .50 2,949 6.54 2,617 6.59 0.05 4,070 10.74 4.15 .50 to .74 1,662 3.69 2,396 6.04 0.00 .7 5 to .99 2,313 5.13 2,038 5.13 0.00 .50 to .99 3,975 8.82 4,434 11.17 2.35 4,942 13.05 1.88 Under 1.00 6,924 15.36 7,051 17.76 2.41 9,012 23.79 6.03 1.00 to 1.24 2,502 5.55 2,237 5.64 0.09 2,302 6.08 0.44 1.25 to 1.49 2,408 5.34 2,522 6.35 1.01 2,34 5 6.19 0.16 1.50 to 1.74 2,627 5.83 2,228 5.61 0.00 1.75 to 1.84 1,063 2.36 961 2.42 0.00 1.50 to 1.84 3,690 8.19 3,189 8.03 0.15 3,072 8.11 0.08 1.85 to 1.99 2,067 4.59 1,384 3.49 1.10 1,088 2.87 0.61 2.00 and over 27,490 60.98 23,309 58.72 2.25 20,060 52.96 5.77 1.00 and over 38,157 84.64 32,641 82.24 2.41 28,867 76.21 6.03 Total Pop. 45,081 39,692 11.95 37,879 4.57 Source: U.S. Bureau of the Census 1990 Census of Population and Housing 2000 Census of Population and Housing 2 005 2009 American Community Survey

PAGE 113

113 CHAPTER 5 FINDINGS AND CONCLUSION S As concerns over taxes on personal property constantly grow, alternatives to traditional property tax must be explored. With this in mind, this research aims to show how the expected theoretical effects of land value taxation are reflected in actual observations before, during, and after implementation. The study here does not concern it self with the accuracy of theories such as those proffered by the likes of Mr. Vickrey in Chapter 1 but it is concerned with how well theory is manifested in reality. In order to assess the virtues of land value taxation as an alternative to traditional property tax, varying metrics of analysis must be considered. This thesis considered how the effec ts of land value taxation manifest themselves in United States ( U S ) Census and American Community Survey (ACS) data. As a result, the hypothesis of this research was as follows: The theoretical effects of land value taxation on income and housing charac teristic data are evident in U S Census and American Community Survey figures When laying out theoretically expected results of the effect of land value taxation for selected income and housing characteristics and comparing them to observed values from A ltoona, Pittsburgh, and Wilkes Barre, PA, it is not overtly evident that land value taxation has left an impact. Observing data from 1990 to 2009 of the ten variables examined only in occupancy status do theoretic al effects present themselves. Despite oc cupancy status display ing the theoretically expected effects of the policy between 2000 and 2009, it is unclear whether or not the policy prompted those observations. The theoretical expectations took into account assumptions about the strength of Bourass

PAGE 114

114 extraneous factors such as local legislation and local housing market characteristics may have played a role in occupancy status observations. Such factors will be further discussed moment arily. Evidence of theoretical effects on sev eral of the variables was unclear, especially for examination of data from 1990 to 2000. This was due to a lack of land value taxation policy intervention between 1990 and 2000. The result of this research is not surprising. Housing and income v ariables such as these with data from United States ( U S ) Cens us es and American Community Survey ( ACS ) are impacted by a myriad of factors in additional to the presence of land value taxation. Among these extraneous f actors were scope of the variables, local legislation, local housing market characteristics, national housing market trends, and shifting employment and population bases. First, time, access, and funding limitations prevented the research from incorporati ng data more narrow in scope The examination of income and housing characteristics from U.S. Census and ACS data was susceptible to numerous extraneous factors. For theoretical results to manifest themselves in variables so broad in scope, the magnitude need to be tremendously significant. Even the variables most likely to be affected by the presence of land value taxation, density measures and units in structure, failed to display evidence of theoretica l expectations. As was seen in Chapter 4, local legislation, such as the tax abatement policies in Pittsburgh, likely played a significant role between land value taxation policy and the housing and income observations. The local housing market character istics of each

PAGE 115

115 study area influenced housing characteristic data as well Local hou sing unit booms in Pittsburgh during the 19 50s and in Wilkes Barre during the 1970s likely made the effects of land value taxation difficult to discern. Broader economic trends also played a role in the assessment of the effects of the tax policy. National trends in the housing market, such as the housing construction boom of the early to 2000s likely prompted an increase in single family units throughout the study areas. More recently, the resulting housing crisis may have also impacted occupancy rates. Shifting population and employment bases at the national level likely played a role in both housing and income characteristics. The Pennsylvania region continued to ble ed manufacturing and other labor intensive jobs and experienced growth in white collar employment sectors. With many, lower skilled jobs being replaced with fewer, higher skills jobs, housing and income characteristics of all three study areas were subject ed to market forces that were unassociated with the presence of land value taxation. With the results of this before and after analysis failing to match theoretically expected results, it is clear that problems exist with the theory, the case studies, or the analysis itself. Much of the theory surrounding land value taxation is grounded in logical economic analysis. Therefore, this research considers the theory presented in the literature review beyond reproach. As was discussed in the methodology, ide al case study selection was thought to have been achieved. However, the prevalence of tax obscured the impact of the c taxation policy. Lastly, it is in analysis method that this research could ha ve been improved. As a result of already described limitations, ideal data for the case study

PAGE 116

116 areas pertaining to theory was unavailable for this research. The distant relationship of several of the variables to the theoretical effects of land value taxa tion provided avenues for extraneous circumstances to interfere with the prescribed virtues of the tax policy. Under ideal circumstances, a study of the effects of land value taxation would be slightly different from this research. First, as was just ment ioned, Pittsburgh would need to be removed as a case study area due to the prevalence its property tax abatement policy. Potential replacements for Pittsburgh as a case study include Coatesville, Connellsville, and Oil City, PA as all three rescinded land value taxation between 2000 and 2009. Secondly, selection of variables closely related to land value taxation would help eliminate some of the influence of extraneous variables such as shifting market and demographic trends. The methodology chapter ment ions that the original intention of this research of to analyze the effect of land value taxation on property values. Were such data available, the limitations of this research would have been reduced, allowing a better way to prove the effects of land va lue taxation. As was discussed in Chapter 2, the lofty hopes and alleged benefits of land value taxation have been enough to persuade several municipalities in the United States and around the world to adopt such policy. While it has not been a fruitful e ndeavor in all attempts, several municipalities and governments both domestically and abroad continue to implement the policy and swear by its virtues. However, the results of this research show that those virtues can sometimes go unfulfilled. Implication s for Planning Netzer (1998c) states that land value taxation is especially applicable in certain municipalities. These municipalities have local governments that are responsible for the

PAGE 117

117 provision of infrastructure projects and public services. Such loca l governments in which land value taxation may be of particular interest are also those that must generate revenue for the aforementioned infrastructure projects and public services on their own. Depending on your level of faith in the theoretical effect s of land value taxation, municipalities. In general, property tax policy influences the physical, economic, and societal structure of settlements everywhere. Planners must be aware of the ramifications of such policy. Not only is property tax a primary source of revenue for local governments and thus of municipal planning departments, but it also influences development patterns, which effect everything from traffic patterns t o infrastructure layout to public services provision to growth management. It also has an impact on inc ome and wealth distribution as well as environmental protection in the way that such policy affects urban sprawl development. As a result, land value t axation and property tax policy in general should not be taken lightly or dismissed as merely an economic development tool by planners. As planners and public officials look for ways to stimulate economic development and facilitate the efficient functionin g of human s ettlement, land value taxation is a course to be considered. Lauded for its ability to increase wealth, fairly distribute wealth, spur investment, preserve the natural environment, and allow for the efficient provision of public goods and infr merit. However, it is also important to note that, as with any complex issue, there is no As this

PAGE 118

118 research shows, the l auded virtues of any economic theory do not always manifest themselves in reality. Future Research Anderson (2009) notes that the opportunities for future research of land value taxation are limited by the relative small amount of communities implementing it in the United States. The City of Pittsburgh, PA was the one major metropolitan area implementing land value taxation before rescinding it in 2001. While leaving the opportunity for before and after intervention study, post land value taxation data in Pittsburgh is muddled by a vast array of property tax abatements that have replaced the policy. Perhaps examining how well the tax abatements have effectively replaced land value taxation in Pittsburgh could be a route of study in the future. Another opp ortunity to extend this research lies in regression analysis. With proper data and a proper model, a more quantitative study could be performed to examine the statistical significance of the effects land value taxation on housing and income characteristic s in the study areas. Geographically, it may be advantageous to study international cases. While studies domestically have dealt with a transition from a pre existing property tax policy to one of land value taxation, burgeoning third world nations may be implementing p roperty tax for the first time (Anderson, 2009) Research on the observed or anticipated effects of land value taxation in these geographies may provide insight into Another dimension of future research to consider is the aspect of land value (1992) which examines a number of municipalities in the Melbourne, Australia area that

PAGE 119

119 have employed the land value tax for varying numbers of years and determines that while espoused advantages of land value taxation persist, they may diminish with time. Such methodology could be applied to instances of the policy in the United States as well. From a depth of an alysis perspective, land value taxation research has often focused on the variables most directly impacted by the policy such as construction activity, land intensity, and timing of development. However, with advent of more statistically sophisticated co mputer software, regression models can be used to provide researchers with opportunities to examine impact on other variables less directly related to property tax such as environmental and quality of life measures. Despite the limited amount of cases to e xamine, research of this topic can still easily be more fully developed as a dissertation or as government research. While a more comprehensive study of land value taxation will likely demand more precise measures, U S Census and ACS data are still usefu l in depicting large scale ramifications of policy interventions at the municipal scale. As a result, within this vein there are plenty of research opportunities to be had that would enhance the literature of land value taxation.

PAGE 120

120 APPENDI X DATA REVIEW & ECONOMIC PROFILES This appendix will detail analysis of income, housing and general economic data pertaining to the three case study areas. From there, conclusions about the impact and effectiveness of land value taxation will be found. The section will be gin with analysis of 1990 and 2000 United States ( U S ) Census d ata and 2005 2009 Ame rican Communities Survey (ACS) d ata. This data will be categorized into income characteristic data and housing characteristic data. The U S Census and ACS data section is focused on the characteristics with changes exhibited by each study area discussed within. Next, discussion will move to Bureau of Labor Statistics (BLS) data, covering total employment figures, location quotients, shift share analysis, Esteban Marquil las Extension analysis, and market forecasts of Altoona, Pittsburgh, and Wilkes Barre. The BLS data section focuses on the metropolitan statistical areas (MSAs) of the three study areas individually, deriving analysis from the tables on a case basis. Fina lly, the section will conclude with a summary of the data from each of the aforementioned sources. This synopsis of data will make synthesizing the analysis U S Census and American Community Survey Analysis Th e first part of this section S Census data from the 1990 and 2000 U S Cens us es and the 2005 2009 American Communities Survey. This data provides sufficient cross sectional insight into the changes in income and housing characteri stics in the three case study areas at three points over the last two decades.

PAGE 121

121 With knowledge of tax policy adoption and removal dates in hand, potential correlations between policy intervention and characteristic changes can be determined. Housing Figure s Land and construction markets are the first to be impacted by shifts in property tax rates. Together, these markets yield figures pertaining to housing characteristics. The following tables and narratives provide insight into what changes have taken pl ace in seven housing related categories at time cross sections of 1990, 2000, and 2009. Total housing units Total housing units decreased in all three study areas between 1990 and 2000. Altoona saw the most decrease followed by Pittsburgh then Wilkes Bar re. Between 2000 and 2009, Pittsburgh saw an increase in total housing units by just over 1% while Wilkes ell by 1.5%. Refer to Table 4 1 Total Housing Units for data. Occupancy status In Altoona, the amount of occupied housing units per total housing units increased by nearly 1.4% between 1990 and 2000 while it fell by over 5% in Pittsburgh and over 2% in Wilkes Barre. Between 2000 and 2009, the occupied to total unit ratio fell in all three study areas. Th e number fell by just over 2% in Altoona, by 2.35% in Wilkes Barre, and by just over 4% in Pittsburgh. See Table 4 2 Occupancy Status for figures. Units in s tructure Between 1990 and 2000, Altoona saw little change in its units per structure makeup with t he exceptions of the 50 unit or more structures category which increased remained unchanged except that single unit detached structures increased by 2.64%

PAGE 122

122 and single uni t attached structures decreased by 1.45%. Wilkes Barre also experienced an increase in single unit detached structures and a decrease in single unit attached structures in addition to an increase in two unit structures and a decrease in tructure category. Between 2000 and 2009, Altoona remained virtually unchanged with a slight increase in the amount of single unit detached homes. Pittsburgh also essentially maintained its housing unit makeup with only an increase in the amount of single unit attached structures. Wilkes volatility with an increase in single unit attached structures of nearly 5% and decreases Refer to Table 4 3 Units in Structure for data. Value of o wner o ccupied housing u nits Between 1990 and 2000, owner occupied housing unit market composition in the three study areas moved similarly. The housing unit market shifted as values for units rose from a predominantly belo w $35,000 to the $50,000 to $150,000 range. In particular, Altoona saw a drastic reduction in the amount of units valued at $25,000 or less. While all markets continued to move in the same increased price direction between 2000 and 2009, increased variati occupied housing unit market was clearly dichotomous with increases observed in every category $80,000 and above and decreases observed in every category $70,000 and below. Altoona also saw decreases in every cate gory $70,000 and below. Increases in the amount of units between $70,000 and $200,000 were observed but growth in higher valued units was less vigorous than in Pittsburgh. The market for owner

PAGE 123

123 occupied housing units in Wilkes Barre shifted less than in P ittsburgh and Altoona. However, decreases in units were observed between $35,000 and $60,000 in Wilkes Barre while units valued between $90,000 and $175,000 saw increases. For more data, see Table 4 4 Value o f Owner Occupied Housing Units. Gross r ents Ove r the past 20 years, rents in Altoona, Wilkes Barre, and Pittsburgh have all moved in the same direction in terms of gross rents in renter occupied housing units rising. Renting in Pittsburgh has been most expensive among the three study areas, with Altoo na maintaining a slightly more affordable renters market than Wilkes Barre. Between 2000 and 2009, Pittsburgh also saw a sharp increase of 13% in $1000/month gross rent units, compared to just below 4% in Altoona and Wilkes B arre. Please see Table 4 5 Gr oss Rents for data. Year b uilt During the 1990s, Altoona saw a sharper drop in the amount of pre 1940 housing units than the other two case studies, falling by about 6.3% versus 4.6% in Pittsburgh and 4% in Wilkes Barre. New unit construction was also the most vigorous in Altoona with 2000 U S Census data showing that 3.36% of units in the city were constructed within the previous decade, compared to 2.35% in Pittsburgh and 1.24% in Wilkes Barre. Units constructed from 1940 to 1970 saw increases in their total housing during the 1990s. Altoona and Wilkes Barre also witnessed decreases in 1970s era housing while Pittsburgh saw a decrease in 1980s housing. The first decade of the 21 st century produced significantly reduced new constru ction growth in Altoona and almost no growth in Wilkes Barre, while Pittsburgh growth was approximately equal to that of the previous decade. The percentage of

PAGE 124

124 units built in Altoona between 2000 and 2009 consisted of 1.51% of the total versus 2.52% in Pi ttsburgh and only 0.40% in Wilkes Barre. Altoona experienced continued decline in pre 1940s housing while the other two study areas had increased their percentages of pre 1940s housing, although it did not drop as significantly as it did during the previo us decade. Age of housing stock was also less volatile in Altoona versus the other 2 case studies, with only 1960s and pre 1940 era housing changing its share by more than 1%. Pittsburgh saw a 2.36% increase in pre 1940 housing and decreases in 1940s and 1960s era housing by 3.24% and 2.21%, respectively. Wilkes Barre experienced increases in pre 1940s and 1980s era housing of 4.23% and 1.22%, respectively. It also saw decreases in 1940s and 1970s era housing of 2.11% and 3.86% respectively. Overall, th e age makeup of housing stock in Altoona is more evenly distributed than in Pittsburgh or Wilkes Barre. Pittsburgh saw a boom in housing unit construction during the 1950s and Wilkes Barre saw the same in the 1970s. These booms are still evident in 2009 ACS data with a higher percentage of 1950s housing than from any other decade since 1940 in Pittsburgh and a greater share of housing from the 1970s and less volatile. Despite a building boom in the 1970s, similar to Wilkes Barre, 2009 data shows precipitous growth in share of housing stock looking further back in time. This is unique among the three case studies. Also unique to Altoona is the continued drop in the share pre 1940s housing stock. During the troubled economic times of the 2000s, Pittsburgh and Wilkes Barre experienced increases in this category. Refer to Table 4 6 Year Structure Built for data.

PAGE 125

125 Density m easures During the 1990s, all three study areas experienc ed drops in both unit and population density. Pittsburgh and Wilkes Barre saw their population densities fall by about 4% compared to just over 2% in Wilkes Barre. Between 2000 and 2009, all three cities experienced different changes in density respectively. It was the only area to experience a substantial drop in unit density. Pittsburgh also saw a decrease in population density of around 6%, but had an increase of 1.18% in unit density. Pittsburgh was the only study area to have an increase in unit density over the decade. Wilkes Barre remained relatively stable over the decade, experiencing a l oss in population density of 4.73% and remaining constant in unit density, falling by only 0.24%. Overall, Altoona saw its density fall the most over the 20 year period. Pittsburgh appeared to recapture unit density growth during the 2000s while unit dens ity in Wilkes Barre appears to have stabilized. All three study areas continue to experience population density decline, however that rate of density decline between in 2000 and 2009 increased only in Altoona. For data associated with the previously ment ioned dens ity measures, please see Table 4 7 Density Measures. Income Figures changes, money is a significant factor. In considering land value taxation, effects on incom e are somewhat removed from direct intervention. Effects on land and

PAGE 126

126 improvements more directly related to the shift from traditional property tax. It is for this reason that data analysis begins with income characteristics. Household Income All three st udy areas saw roughly the same growth in median household income between 1990 and 2000. However, between 2000 and 2009, Pittsburgh maintained a 25% rate of growth, Altoona held a 19% rate, and Wilkes growth fell to 7.5%. See Table 4 8 Household I ncomes and Table 4 9 Media n Household Income for figures. Per Capita Income Per capita income tells a similar story, with all three study areas having growth rates between 43% and 50% during the 1990s but with Pittsburgh maintaining the highest rate during the past decade at 30%, followed by Altoona at 20%, and W ilkes Barre at 14%. See Table 4 10 Per Capita Income for data. Ratio of Income to Poverty Level depth of poverty. Rati os below 1.0 indicate poverty, with ratios between 1.25 and 1.0 Between 1990 and 2000, Altoona saw little change in its income to poverty ratios. lightly as ratios below 1.0 fell. Wilkes Barre saw the opposite as ratios below 1.0 increased by nearly 2.5%. All three study areas saw little change in its near pove rty and severe poverty figures. Between 2000 and 2009, Altoona and Pittsburgh saw their below poverty ratios increase by about 1.25%, but Wilkes Near poverty ratios remained virtually unchanged in all three study areas. However,

PAGE 127

127 Altoona saw its severe poverty ratios fall by nearly 2% while Pitt constant and Wilkes See Table 4 11 Ratio of Incom e to Poverty Level for figures. Bureau of Labor Statistics Data Analysis For part two of this cross sectional case study analysis, Bureau of Labor Statistics data will be analyzed to better understand changes in the local economy of the MSAs of the case studies between 2001 and 2010. Data from the Altoona, PA MSA, Pittsburgh, PA MSA, and Scranton Wilkes Barre, PA MSA will be used in this analysis. From this d ata, tables describing employment changes by industry, location quotients, Shift Share analysis, and Esteban Marquillas Extension have been developed to assist in developing a narrative. These tables can be found immediately following the narrative of the three case study areas. Altoona, PA MSA: Between 2001 and 2010 Altoona gained over 1250 jobs relative to its share of national employment despite losing 143 total jobs overall. This means that the economy of Altoona shrank over the period but not as much as that of the national economy approximately 32 jobs. The health care and social assistance sector was a rapidly growing industry nationally and Altoona had a very sizable amount of employment there. Accommodation and food services and professional and technical services were other industries of significant employment in Altoona that experienced growth nationally. and construction sectors that experienced slow growth over the period. As a result, these slow growth sectors negated much of the gains of the high growth sectors.

PAGE 128

128 Competitive shift figures show that Altoona was a highly competitive environment for the m anufacturing, transportation and warehousing, retail trade, and administrative and waste services industries. Each of these industries grew more rapidly in the local area than nationwide. These locally high performing industries helped to offset some of The shift share figures also show that Altoona remains strong in the transportation and warehousing sector. Historically a railroad town, Altoona may capitalize on rising fuel costs and once a gain utilize this industry to catalyze future growth across a myriad of industrie s within the metropolitan area. Sectors of future growth in the Altoona, PA MSA: Heath care and social assistance: Despite having a less than proportional share of employment within it here than in the rest of the country, health care and social assistance is a sector with potential for Competitive shift for the sector was positive and Esteban Marquillas Extension showed it was highly competitive in the area and containing comparative advantage, making it an industry worth intervening upon. Like education services, health care also utilizes a large proportion of local in puts and is an export i ndustry. Transportation and warehousing: Shift share analysis shows that this sector is very competitive in the local marketplace. Esteban Marquillas Extension illustrates that the industry is specialized locally, but lacks comparative advantage, opening the possibility for future intervention Philadelphia and Pittsburgh, Altoona has long been a focal point for transportation and

PAGE 129

129 logistics in Pennsylvania. Traditionally a pil lar of the local economy, transportation and warehousing employment increased from 2001 to 2010 but still has less of a share of the local market than nationwide according to location quotient figures. As a result, local expansion of this industry would s eem likely. Administration, support and waste management services: Local employment in this sector grew by over 30% over the period and exhibited a high competitive component in Shift Share Analysis. A decline in the location quotient however means that t share nationally. The sector shows comparative advantage but a lack of specialization locally according to Esteban Marquillas Extension data, pointing toward a possible intervention in po licy to help spur this industry. Policy makers in Altoona may find it fruitful to expand this well performing industry by reaching out to larger markets in corners of the Commonwealth made possible by a strong transportation and may help ease the office and facility management burden of co mpanies from around the region. Pittsburgh, PA M SA: Despite economic revival efforts, Pittsburgh lost a total of 33,000 jobs during the period and lost nearly 5500 jobs relative to its share of the national market. The Steel City MSA saw substantial growth in health care and education over the period, y et experienced losses in the information, transportation and warehousing, and construction sectors along with continued significant decline in the manufacturing industry.

PAGE 130

130 The industrial mix of Pittsburgh, on the other hand, was a positive factor in job g rowth, by itself creating approximately 24,000 jobs. Industries such as health care, education, and accommodation and food services, in fact growing significantly in Pi ttsburgh, grew nationally. The c ity benefited from having concentrated employment in th ese sectors. Conversely, continued reliance on manufacturing resulted in tremendous overall job losses as this sector continues to decline nationally. However, competitive shift was extraordinarily negative resulting in a loss of over 45,000 jobs in total This was particularly the case in the health care and social assistance sector, whose competitive shift amounted to over 14,000 alone. Evidently, that sector in Pittsburgh was significantly outperformed nationally. Esteban Marquillas Extension shows a pair of efficiently functioning sectors: finance and insurance and arts, entertainment, and recreation. These industries exhibited both specialization and comparative advantage over the period. Sectors of future growth for the Pittsburgh, PA MSA: Constr uction: This sector was locally competitive over the period as it outpaced the national construction industry according to competitive shift figures. It also exhibited comparative advantage but not specialization, creating the possibility for effective po licy intervention. Capitalizing on the construction multipliers of its reemerging downtown will Manufacturing: manufacturing indu stry. Today, it maintains a high location quotient in the area as well as strong competitive shift numbers over the period according to shift share analysis. A

PAGE 131

131 lack of specialization or comparative advantage according to Esteban Marquillas Extension poin ts away from policy intervention as the industry may be unsustainable. However, the city is already invested in the sector and, in conjunction with burgeoning construction and education sectors, it may be settling into an appropriate share of the local ec onomy. Education: Despite having two renowned universities in the University of Pittsburgh and the Carnegie Mellon Institute, Pittsburgh maintains a low location quotient in education. Shift share analysis shows that this sector suffered in competitive sh ift over the period. Esteban Marquillas Extension reveals that the sector exhibits specialization but lacks comparative advantage. Policy intervention to stimulate the education industry will have the collateral effects of augmenting the already well fun ctioning arts and entertainment and finance and insurance sectors as well as other skill intensive industries such as health care and professional, technical, and scientific services. Health care and social assistance: The Steel City also has a below natio nal share location quotient in this industry. While exhibiting specialization according to the Esteban Marquillas Extension, health care lacks comparative advantage, something that could potentially be improved with more focus on the educational services sector. Scranton Wilkes Barre, PA MSA: Over the period, Wilkes Barre lost over 7300 jobs and lost over 1000 jobs relative to its share of national employment. The area saw gains in the business management, transportation and warehousing, arts and entertai nment, and health care sectors, but continued to experience a mass exodus of manufacturing jobs. The sector has now

PAGE 132

132 been replaced as the sector of highest employment in Wilkes Barre by another industry that showed decline over the period in retail trade. Like Altoona and Pittsburgh, Wilkes Barre has a strong mix component according to shift share analysis. By itself, the mix contributed a gain of 250 jobs over the period. Health care, education, and accommodation and food services were all sectors that sh owed high growth nationally and also held significant employment in Wilkes Barre. However, local commitment to sectors that grew slowly over the period such as manufacturing, construction and information counteracted many of those gains. Competitive shift analysis shows local strengths particularly in transportation and warehousing, management of companies and enterprises, administrative and waste services, and arts, entertainment, and recreation. Locally, these sectors grew more rapidly than at the natio nal level. However, several other industries such as manufacturing, health care, and educational services, performed below national standards, resulting in significant local jobs losses amounting to nearly 2400 due to lack of local competitiveness. Sector s of future growth in the Scranton Wilkes Barre, PA MSA: Transportation and warehousing: According to Esteban Marquillas Extension, the Scranton Wilkes Barre MSA exhibited only one well functioning industry over the period: transportation and warehousing. Locally, the industry was specialized and offered comparative advantage. However, location quotient figures show that the sector maintained less of a presence in the area than would be expected from national averages. In fact, the location quotient for transportation and warehousing in Scranton Wilkes Barre actually fell from 0.8 in

PAGE 133

133 2001 to 0.56 in 2010. For this reason, policy interventions could increase the presence of this successful and efficient local industry. Management of companies and enterpri ses: Over the period, employment in the management of companies and enterprises sector more than doubled. Perhaps it was the draw of the scenic Pocono Mountains, but the Scranton Wilkes Barre area seemed to attract executives more than it ever had before. Shift share analysis shows that the sector was quite strong locally, exhibiting a positive competitive shift figure. Esteban Marquillas Extension also shows that the industry has had comparative advantage, but has lacked specialization. This market fla w could potentially be remedied with policy intervention and could make the business managem ent sector even more prevalent. Accommodation and food services: The accommodation and food services sector experienced modest gains over the period, increasing emp loyment by about 5% and exhibit a strong location quotient for the area. Shift share analysis also shows this sector to yield positive competitive shift figures, revealing a locally efficient market in this industry. While Esteban Marquillas Extension do es not show policy intervention to be advisable, capitalization on the Poconos tourism industry could prove fruitful. The Scranton Wilkes position in the Poconos and adjacent to the megalopolis of the northeastern United States in conjunction with growing transportation and warehousing and business management industries could demand more from this industry in the future.

PAGE 134

134 BLS Analysis Conclusions In evaluating the case study economic profiles developed from the BLS data sets, differences and si milarities between the study areas become evident. However, the similarities outweigh the differences in all three cases. The MSAs representing each case study area produced their own employment idiosyncrasies. Altoona was the only study area to maintain a strong competitive shift according to shift share analysis, which led to it being the only study area to experience the other two cases over the period, growing by a modest 3.48% compared to 10.06% and 4.16% in Pittsburgh and Wilkes Barre, respectively. Pittsburgh and Wilkes Barre had weak competitive shift leading to a reduction in their shares of national employment. Table A 1 Employment Figures by Industry, sho economy was a more diverse workplace as it had the lowest percent of employment in manufacturing, retail, and health care of the three study areas. This diversity likely is associated with the sheer size of Pittsburgh in comparison to the two other cases. While Wilkes Barre has experienced massive overall job loss, the management of enterprises and companies sector was a high performer, producing positive indicators in competitive shift and overall job growth. The significance of the se differences among the economies of the study areas show that while quite similar, the individual metropolitan economies of Pennsylvania cannot be painted with a wide brush. Despite the aforementioned unique changes to the economies of the three case stu dies, several shifts in all three show how closely related the three economies are. First, all three areas experience job loss between 2001 and 2010. Altoona faired best, only losing 0.29% of its 2001 job total. According to shift share analysis, all th ree study

PAGE 135

135 areas maintained a strong industrial mix component meaning that the three areas had significant employment in sectors that experienced high growth nationally, thus spelling growth locally. All three areas have had and continue to have a persiste nt reliance on the manufacturing sector, a staple of the rust belt economy. However, this reliance diminished across the study areas between 2001 and 2010. All three areas experienced increased employment in health care and education over the period. Co nversely, however, all three areas also experienced a weak health care and social assistance sector competitive shift component of job growth over the period, meaning that the industry locally performed below the national average. With an eye toward the f uture, the economies of all three areas will likely depend on increasing local competitiveness in this sector. Additional sectors to take note of for future growth in all three areas are transportation and warehousing, accommodations and food services, an d even an evolving manufacturing sector. As a result, extraneous circumstances emanating from economic growth that may have affected income and housing characteristics are less than likely to have impacted figures in a substantial way. With the sole exce ption of retail trade, significant sectors of employment in the three study areas shifted in a synchronized fashion. Were employment in an already significant industry to suddenly sky rocket, one would expect such economic growth to have an impact on seve ral metrics of income and housing. However, as it stands, there is little to suggest that any abrupt shifts occurred in the local economies of the study areas that could have conclusively affected income and housing characteristics in Altoona, Pittsburgh, or Wilkes Barre.

PAGE 136

136 Summary In this section U S Census and ACS data from 1990, 2000, and 2009 has been used to produce a cross sectional study of income and housing characteristics in Altoona, Pittsburgh and Wilkes Barre, PA. Additionally, BLS data from 2 001 and 2010 was used to depict economic conditions in the MSAs of Altoona, Pittsburgh, and Scranton Wilkes Barre, PA. Together, this triumvirate of analysis summarizes changes rvention. Altoona experienced a decrease in severe poverty between 2000 and 2009 according to income to poverty ratio figures. Units in structure composition rem ained largely unchanged in the c ity and cheap, affordable housing and unit and population dens ity declined. Altoona improved its share of employment relative to the national economy and has witnessed a renewed significance of the transportation and warehousing industry related to the local economy. Pittsburgh maintained the highest rates of income growth over the 20 year period, but also left poverty levels unchanged. Occupancy rates fell continuously throughout the period while the construction of new units became the most vigorous of the study areas between 2000 and 2009. In particular, high en d renter occupied and owner occupied units experienced significant increases during the first decade of the 21 st century in the Steel City. Units in structure composition maintained its 1990 levels while population density fell but at a decreasing rate. Curiously, unit density increased from 2000 2009 while population density decreased. The economic climate in Pittsburgh has been muddled over the past 10 years, with employment in the health care and education sectors increasing despite negative competiti ve shift. Also complicating

PAGE 137

137 matters is the continued decline in manufacturing employment despite strong competitive shift figures. Wilkes Barre has generally struggled across all three categories of data analysis. Income growth fell the most drastically b etween 2000 and 2009 and both poverty and severe poverty increased. In housing analysis, occupancy rates fell over the 20 year period. As units in structure makeup shifted to less dense housing, overall units in structure makeup experienced increased vol atility. Conversely, the value of owner occupied housing experienced the least volatility of the three study areas. Growth in new unit construction slowed to near non existence between 2000 and 2009, prompting a sharp increase in the percentage of pre 19 40 housing. Population density fell throughout the period but at a lesser rate over the final ten years, the least of the three study areas. Data indicates that population and unit densities may be stabilizing. Lastly, the Scranton Wilkes are of national employment fell between 2001 and 2010 thanks, in part, to its reliance on a declining manufacturing sector. However, growth and positively competitive shift in the sectors of transportation and warehousing and management of companies and e nterprises may lead to future growth in employment.

PAGE 138

138 Table A 1. Employment figures by i ndustry Altoona, PA MSA Industry 2001 % 2010 % % Change NAICS 11 Agriculture, forestry, fishing and hunting ND NC ND NC NC NAICS 21 Mining, quarrying, and oil an d gas extraction ND NC ND NC NC NAICS 22 Utilities 213 0.43 144 0.29 32.39 NAICS 23 Construction 2,611 5.26 2,271 4.59 13.02 NAICS 31 33 Manufacturing 9,504 19.16 7,413 14.99 22.00 NAICS 42 Wholesale trade 2,988 6.02 2,362 4.78 20.95 NAICS 44 45 R etail trade 8,275 16.68 8,563 17.31 3.48 NAICS 48 49 Transportation and warehousing 2,095 4.22 2,865 5.79 36.75 NAICS 51 Information 1,347 2.72 851 1.72 36.82 NAICS 52 Finance and insurance 1,468 2.96 1,326 2.68 9.67 NAICS 53 Real estate and rental a nd leasing 389 0.78 325 0.66 16.45 NAICS 54 Professional and technical services 1,896 3.82 1,957 3.96 3.22 NAICS 55 Management of companies and enterprises 640 1.29 467 0.94 27.03 NAICS 56 Administrative and waste services 1,591 3.21 2,110 4.27 32.62 NAICS 61 Educational services 370 0.75 446 0.90 20.54 NAICS 62 Health care and social assistance 8,825 17.79 10,513 21.26 19.13 NAICS 71 Arts, entertainment, and recreation 758 1.53 822 1.66 8.44 NAICS 72 Accommodation and food services 4,386 8.84 4,59 0 9.28 4.65 NAICS 81 Other services, except public administration 1,843 3.72 1,971 3.99 6.95 NAICS 99 Unclassified 2 0.00 NC NC NC Base Industry: Total, all industries 49,599 100.00 49,456 100.00 0.29

PAGE 139

139 Table A 1. Continued Pittsburgh, PA MSA Industry 2001 % 2010 % % Change NAICS 11 Agriculture, forestry, fishing and hunting ND NC ND NC NC NAICS 21 Mining, quarrying, and oil and gas extraction ND NC ND NC NC NAICS 22 Utilities 8,998 0.92 5,626 0.60 37.47 NAICS 23 Construction 57,111 5.84 4 6,764 4.95 18.12 NAICS 31 33 Manufacturing 124,016 12.68 87,477 9.26 29.46 NAICS 42 Wholesale trade 42,349 4.33 39,829 4.22 5.95 NAICS 44 45 Retail trade 139,899 14.31 125,829 13.32 10.06 NAICS 48 49 Transportation and warehousing 42,906 4.39 33,70 9 3.57 21.44 NAICS 51 Information 26,145 2.67 17,534 1.86 32.94 NAICS 52 Finance and insurance 52,803 5.40 53,704 5.68 1.71 NAICS 53 Real estate and rental and leasing 13,742 1.41 12,783 1.35 6.98 NAICS 54 Professional and technical services 63,692 6.51 65,845 6.97 3.38 NAICS 55 Management of companies and enterprises 14,488 1.48 ND NC NC NAICS 56 Administrative and waste services 53,914 5.51 ND NC NC NAICS 61 Educational services 34,503 3.53 40,313 4.27 16.84 NAICS 62 Health care and social assi stance 155,209 15.87 179,485 19.00 15.64 NAICS 71 Arts, entertainment, and recreation 16,376 1.67 18,886 2.00 15.33 NAICS 72 Accommodation and food services 82,845 8.47 88,027 9.32 6.26 NAICS 81 Other services, except public administration 42,028 4.30 3 8,733 4.10 7.84 NAICS 99 Unclassified 178 0.02 NC NC NC Base Industry: Total, all industries 977,906 100.00 944,813 100.00 3.38

PAGE 140

140 Table A 1. Continued Scranton -Wilkes Barre, PA MSA Industry 2001 % 2010 % % Change NAICS 11 Agriculture, forestry fishing and hunting ND NC ND NC NC NAICS 21 Mining, quarrying, and oil and gas extraction ND NC ND NC NC NAICS 22 Utilities 2,785 1.27 ND NC NC NAICS 23 Construction 9,592 4.36 8,272 3.89 13.76 NAICS 31 33 Manufacturing 42,879 19.50 27,603 12.98 35 .63 NAICS 42 Wholesale trade 8,593 3.91 ND NC NC NAICS 44 45 Retail trade 33,125 15.06 31,747 14.93 4.16 NAICS 48 49 Transportation and warehousing 10,397 4.73 14,133 6.65 35.93 NAICS 51 Information 6,719 3.06 4,945 2.33 26.40 NAICS 52 Finance and i nsurance 11,115 5.05 10,185 4.79 8.37 NAICS 53 Real estate and rental and leasing 2,394 1.09 1,814 0.85 24.23 NAICS 54 Professional and technical services 7,798 3.55 7,183 3.38 7.89 NAICS 55 Management of companies and enterprises 1,490 0.68 3,349 1. 58 124.77 NAICS 56 Administrative and waste services 12,331 5.61 12,944 6.09 4.97 NAICS 61 Educational services 7,134 3.24 7,617 3.58 6.77 NAICS 62 Health care and social assistance 35,578 16.18 41,832 19.68 17.58 NAICS 71 Arts, entertainment, and recr eation 2,047 0.93 3,364 1.58 64.34 NAICS 72 Accommodation and food services 17,321 7.88 18,238 8.58 5.29 NAICS 81 Other services, except public administration 7,744 3.52 6,696 3.15 13.53 NAICS 99 Unclassified 33 0.02 NC NC NC Base Industry: Total, all industries 219,915 100.00 212,608 100.00 3.32

PAGE 141

141 Table A 1. Continued U.S. TOTAL Industry 2001 % 2010 % % Change NAICS 11 Agriculture, forestry, fishing and hunting ND NC ND NC NC NAICS 21 Mining, quarrying, and oil and gas extraction ND NC ND NC N C NAICS 22 Utilities 599,899 0.55 551,195 0.52 8.12 NAICS 23 Construction 6,773,512 6.20 5,489,076 5.17 18.96 NAICS 31 33 Manufacturing 16,386,001 14.99 11,487,828 10.82 29.89 NAICS 42 Wholesale trade 5,730,294 5.24 5,465,985 5.15 4.61 NAICS 44 45 Retail trade 15,179,753 13.89 14,479,851 13.63 4.61 NAICS 48 49 Transportation and warehousing 4,138,146 3.79 3,942,399 3.71 4.73 NAICS 51 Information 3,591,995 3.29 2,704,037 2.55 24.72 NAICS 52 Finance and insurance 5,642,689 5.16 5,486,175 5.17 2.77 NAICS 53 Real estate and rental and leasing 2,036,285 1.86 1,915,361 1.80 5.94 NAICS 54 Professional and technical services 6,871,441 6.29 7,456,791 7.02 8.52 NAICS 55 Management of companies and enterprises 1,716,130 1.57 1,855,073 1.75 8.10 NAI CS 56 Administrative and waste services 7,737,320 7.08 7,395,915 6.96 4.41 NAICS 61 Educational services 1,883,564 1.72 2,464,132 2.32 30.82 NAICS 62 Health care and social assistance 12,966,103 11.86 16,194,487 15.25 24.90 NAICS 71 Arts, entertainment and recreation 1,784,330 1.63 1,903,525 1.79 6.68 NAICS 72 Accommodation and food services 10,100,636 9.24 11,101,250 10.45 9.91 NAICS 81 Other services, except public administration 4,206,345 3.85 4,349,437 4.10 3.40 NAICS 99 Unclassified 254,603 0.2 3 NC NC NC Base Industry: Total, all industries 109,304,802 100.00 106,198,248 100.00 2.84

PAGE 142

142 Table A 2. Location q uotients in s tudy a rea MSAs Altoona, PA MSA Pittsburgh, PA MSA Scranton Wilkes Barre, PA MSA Industry 2001 2010 Change 2001 2010 Chang e 2001 2010 Change Base Industry: Total, all industries 1 1 0 1 1 0 1 1 0 NAICS 11 Agriculture, forestry, fishing and hunting ND ND NC ND 9.5 0 NC 6.54 6.91 + NAICS 21 Mining, quarrying, and oil and gas extraction ND ND NC ND 0.95 NC 2.23 2.18 NAICS 2 2 Utilities 1.28 1.78 + 0.6 0.87 + 0.43 ND NC NAICS 23 Construction 1.18 1.13 1.06 1.04 1.42 1.33 NAICS 31 33 Manufacturing 0.78 0.72 1.18 1.17 0.77 0.83 + NAICS 42 Wholesale trade 0.87 1.08 + 1.21 1.22 + 1.34 ND NC NAICS 44 45 Retail trade 0 .83 0.79 0.97 1.02 + 0.92 0.91 NAICS 48 49 Transportation and warehousing 0.9 0 0.64 0.86 1.04 + 0.8 0 0.56 NAICS 51 Information 1.21 1.48 + 1.23 1.37 + 1.08 1.09 + NAICS 52 Finance and insurance 1.74 1.93 + 0.96 0.91 1.02 1.08 + NAICS 53 Real estate and rental and leasing 2.38 2.74 + 1.33 1.33 0 1.71 2.11 + NAICS 54 Professional and technical services 1.64 1.77 + 0.97 1.01 + 1.77 2.08 + NAICS 55 Management of companies and enterprises 1.22 1.85 + 1.06 ND NC 2.32 1.11 NAICS 56 Administrativ e and waste services 2.21 1.63 1.28 ND NC 1.26 1.14 NAICS 61 Educational services 2.31 2.57 + 0.49 0.54 + 0.53 0.65 + NAICS 62 Health care and social assistance 0.67 0.72 + 0.75 0.8 0 + 0.73 0.78 + NAICS 71 Arts, entertainment, and recreation 1.07 1. 08 + 0.97 0.9 0 1.75 1.13 NAICS 72 Accommodation and food services 1.04 1.13 + 1.09 1.12 + 1.17 1.22 + NAICS 81 Other services, except public administration 1.04 1.03 0.9 1 + 1.09 1.3 0 + NAICS 99 Unclassified 57.77 NC NC 12.8 703.14 + 15.52 NC NC Source: U S Department of Labor, Bureau of Labor Statistics

PAGE 143

143 Table A 3 Shift s hare a nalysis Altoona, PA MSA Pittsburgh, PA MSA NAICS 2 digit code Industry Economic Share Mix Component Competitive Component Economic Share Mix Component Competitiv e Component 11 Agriculture, Forestry, Fishing, and Hunting 0.00 0.00 0.00 0.00 0.00 0.00 21 Mining 0.00 0.00 0.00 0.00 0.00 0.00 22 Utilities 6.05 11.24 51.71 255.73 474.79 2,641.48 23 Construction 74.21 420.91 155.11 1,623.15 9,206.59 482.75 31 33 Manufacturing 270.11 2,570.86 749.98 3,524.66 33,546.73 532.39 42 Wholesale Trade 84.92 52.90 488.18 1,203.60 749.74 566.66 44 45 Retail Trade 235.18 146.36 669.54 3,976.07 2,474.33 7,619.59 48 49 Transportation and Warehous ing 59.54 39.56 869.10 1,219.43 810.15 7,167.41 51 Information 38.28 294.70 163.02 743.07 5,720.10 2,147.83 52 Finance and Insurance 41.72 1.00 101.28 1,500.72 36.09 2,365.62 53 Real estate and rental and leasing 11.06 12.04 40.90 390 .56 425.50 142.94 54 Professional and technical services 53.89 215.40 100.51 1,810.19 7,235.85 3,272.66 55 Management of Companies and enterprises 18.19 70.01 224.82 411.76 1,584.76 0.00 56 Administrative and waste services 45.22 24.98 589.20 1,532.29 846.63 0.00 61 Educational Services 10.52 124.56 38.04 980.61 11,615.42 4,824.81 62 Health Care and Social Assistance 250.82 2,448.12 509.31 4,411.20 43,056.14 14,368.94 71 Arts, Entertainment and recreation 21.54 72.18 13.36 465.4 2 1,559.36 1,416.07 72 Accommodation and Food Services 124.65 559.15 230.50 2,354.54 10,561.53 3,024.99 81 Other Services 52.38 115.08 65.30 1,194.48 2,624.19 4,724.71 99 Unclassified 0.06 0.00 0.00 5.06 0.00 0.00 Total All Industries 1,398.3 4 31.94 1,163.34 27,602.55 24,018.76 45,705.20 Source: U S Department of Labor, Bureau of Labor Statistics

PAGE 144

144 Table A 3. Continued Scranton Wilkes Barre, PA MSA NAICS 2 digit code Industry Economic Share Mix Component Competitive Component 11 Agricu lture, Forestry, Fishing, and Hunting 0.00 0.00 0.00 21 Mining 0.00 0.00 0.00 22 Utilities 79.15 146.95 0.00 23 Construction 272.61 1546.28 498.90 31 33 Manufacturing 1,218.66 11,598.91 2,458.43 42 Wholesale Trade 244.22 152.13 0.00 44 4 5 Retail Trade 941.45 585.87 149.31 48 49 Transportation and Warehousing 295.49 196.32 4,227.81 51 Information 190.96 1,470.01 113.03 52 Finance and Insurance 315.90 7.60 621.70 53 Real estate and rental and leasing 68.04 74.13 437.83 5 4 Professional and technical services 221.63 885.91 1,279.28 55 Management of Companies and enterprises 42.35 162.98 1,738.37 56 Administrative and waste services 350.46 193.64 1,157.10 61 Educational Services 202.76 2,401.66 1,715.90 62 Health Care and Social Assistance 1,011.16 9,869.60 2,604.44 71 Arts, Entertainment and recreation 58.18 194.92 1,180.26 72 Accommodation and Food Services 492.28 2,208.18 798.90 81 Other Services 220.09 483.53 1,311.44 99 Unclassified 0.94 0.00 0.00 Total All Industries 6226.34 250.14 2,389.20 Source: U S Department of Labor, Bureau of Labor Statistics

PAGE 145

145 Table A 4. Esteban Marquillas e xtension Altoona, PA MSA NAICS 2 digit code Industry Homothe tic Employment Specialization Effect Comparative A dvantage Allocation Effect 11 Agriculture, Forestry, Fishing, and Hunting 0.00 0.00 0.00 0.00 21 Mining 0.00 0.00 0.00 0.00 22 Utilities 272.21 59.21 0.24 14.37 23 Construction 3,073.60 462.60 0.06 27.48 31 33 Manufacturing 7,435.44 2,068.56 0.08 163.23 42 Wholesale Trade 2,600.22 387.78 0.16 63.35 44 45 Retail Trade 6,888.08 1,386.92 0.08 112.22 48 49 Transportation and Warehousing 1,877.76 217.24 0.41 90.12 51 Information 1,629.93 282.93 0.12 34.24 52 Finance and Insurance 2,560.47 1,0 92.47 0.07 75.37 53 Real Estate, Rental and Leasing 924.00 535.00 0.11 56.25 54 Professional, Scientific, and Technical Services 3,118.04 1,222.04 0.05 64.78 55 Management of Companies and Enterprises 778.72 138.72 0.35 48.73 56 Administrative, Support and Waste Management and Remediation Services 3,510.95 1,919.95 0.37 711.02 61 Educational Services 854.70 484.70 0.10 49.84 62 Health Care and Social Assistance 5,883.60 2,941.40 0.06 169.75 71 Arts, Entertainment and recreation 809.67 5 1.67 0.02 0.91 72 Accommodation and Food Services 4,583.34 197.34 0.05 10.37 81 Other Services 1,908.70 65.70 0.04 2.33 92 Public Administration 115.53 113.53 0.00 0.00 99 Unclassified 49,599.00 0.00 0.03 0.00

PAGE 146

146 Table A 4. Continued Pittsbu rgh, PA MSA NAICS 2 digit code Industry Homothe tic Employment Specialization Effect Comparative Advantage Allocation Effect 11 Agriculture, Forestry, Fishing, and Hunting 0.00 0.00 0.00 0.00 21 Mining 0.00 0.00 0.00 0.00 22 Utilities 5,367.05 3 ,630.95 0.29 1,065.91 23 Construction 60,599.88 3,488.88 0.01 29.49 31 33 Manufacturing 146,598.95 22,582.95 0.00 96.95 42 Wholesale Trade 51,266.63 8,917.63 0.01 119.32 44 45 Retail Trade 135,807.13 4,091.87 0.05 222.86 48 49 Transp ortation and Warehousing 37,022.32 5,883.68 0.17 982.86 51 Information 32,136.13 5,991.13 0.08 492.18 52 Finance and Insurance 50,482.86 2,320.14 0.04 103.94 53 Real Estate, Rental and Leasing 18,217.82 4,475.82 0.01 46.55 54 Professiona l, Scientific, and Technical Services 61,476.01 2,215.99 0.05 113.86 55 Management of Companies and Enterprises 15,353.52 865.52 0.00 0.00 56 Administrative, Support and Waste Management and Remediation Services 69,222.68 15,308.68 0.00 0.00 6 1 Educational Services 16,851.49 17,651.51 0.14 2,468.34 62 Health Care and Social Assistance 116,002.50 39,206.50 0.09 3,629.66 71 Arts, Entertainment and recreation 15,963.68 412.32 0.09 35.65 72 Accommodation and Food Services 90,366.32 7,521.32 0.04 274.63 81 Other Services 37,632.47 4,395.53 0.11 494.14 92 Public Administration 2,277.83 2,099.83 0.00 0.00 99 Unclassified 977,906.00 0 0.01 0.00

PAGE 147

147 Table A 4. Continued Scranton Wilkes Barre, PA MSA NAICS 2 digit code Ind ustry Homothe tic Employment Specialization Effect Comparative Advantage Allocation Effect 11 Agriculture, Forestry, Fishing, and Hunting 0.00 0.00 0.00 0.00 21 Mining 0.00 0.00 0.00 0.00 22 Utilities 1,206.96 1,578.04 0 0 23 Construction 13,627.92 4,0 35.92 0.05 209.91 31 33 Manufacturing 32,967.70 9,911.30 0.06 568.26 42 Wholesale Trade 11,529.02 2,936.02 0.00 0.00 44 45 Retail Trade 30,540.79 2,584.21 0.00 11.65 48 49 Transportation and Warehousing 8,325.71 2,071.29 0.41 842.26 51 Information 7,226.89 507.89 0.02 8.54 52 Finance and Insurance 11,352.77 237.77 0.06 13.30 53 Real Estate, Rental and Leasing 4,096.89 1,702.89 0.18 311.44 54 Professional, Scientific, and Technical Services 13,824.95 6,026.95 0.16 988.73 55 Management o f Companies and Enterprises 3,452.76 1,962.76 1.17 2,289.92 56 Administrative, Support and Waste Management and Remediation Services 15,567.04 3,236.04 0.09 303.66 61 Educational Services 3,789.62 3,344.38 0.24 804.40 62 Health Care and Social Ass istance 26,087.06 9,490.94 0.07 694.77 71 Arts, Entertainment and recreation 3,589.97 1,542.97 0.58 889.64 72 Accommodation and Food Services 20,321.90 3,000.90 0.05 138.41 81 Other Services 8,462.93 718.93 0.17 121.75 92 Public Administration 512.25 479.25 0.00 0 99 Unclassified 2,19915 0 0.00 0 Source: U S Department of Labor, Bureau of Labor Statistics

PAGE 148

148 Table A 5. List of v ariables for an a ttempted p opulation d en sity r egression m odel Variables Expected Sign Measurement PD, population den sity LVT, land value taxation dummy + O rdinal AR, area I nterval POP, total population + I nterval PMIN, percentage of minority residents + R atio CS, citizenship status in the US O rdinal POV, poverty + I nterval HHI, median household income I nt erval HUT, total housing units + I nterval OS, Occupancy Status + O rdinal MYB, median year structure built I nterval

PAGE 149

149 LIST OF REFERENCES Alonso, W. (1968). "Urban and Regional Imbalances in Economic Development." Economic Development and Cultural Cha nge, 17: 1 14. American Journal of Economics and Sociology. ( 2005 ) Review of Progress and Poverty ([1879] 1938). Henry George. New York: Modern Library. American Journal of Economics and Sociology 64(5): 73 84 Anderson, J. E. ( 1986 ) Regional Science and Urban Economics 14 (4): 483 492 Anderson, J. E. ( 1993 ) Intergovernmental Perspective 19 (3): 19 20, 28. Anderson, J. E. ( 1999 ) Journal of Real Estate Finance and Economics 18 (2):181 190. Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Ri chard W. England. Cambridge, MA: Lincoln Institute of Land Policy, May: 99 126. Arnott, R. J. ( 2005 ) Journal of Public Economic Theory 7 (1):27 50. Blair, J. P., & Carroll, M. (2009). Local economic development: Analysis, prac tices and globalization Thousand Oaks, CA: SAGE Publications, Inc. Blakely, E. J. & Leigh, N. G. (2009). Planning Local Economic Development: Theory and Practice (4th ed.) Thousand Oaks, CA: SAGE Publications, Inc. Bourassa, S. C. ( 1990 ) tion and housing development: Effects of the American Journal of Economics and Sociology 49 (1):101 111. Bourassa, S. C. ( 2009 ) Land Value Taxation, Theory, Evidence and Practice Ed s. Richard F. Dye and Richard W. England. Cambridge, MA: Lincoln Institute of Land Policy, May: 11 25. Brueckner, J.K. ( 1986 ) National Tax Journal 39: 49 58. Brueckner, J. K. & H. A. Kim. ( 2003 ) Urban sprawl and the property tax. International Tax and Property Finance 10 (1): 5 23. Cord, S. ( 1987 ) The evidence for land value taxation Columbia, MD: Center for the Study of Economics.

PAGE 150

150 Daniels, T. ( 2001 ) ing urban growth and The American Journal of Economics and Sociology 60 (1), 229 43. Retrieved January 19, 2010, from OmniFile Full Text Mega database. DiMasi, J. A. ( 1987 ) e National Tax Journal 40 (4): 577 590. Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. En gland. Cambridge, MA: Lincoln Institute of Land Policy, May: 1 10. Edwards, Mary E. (2007). Regional and Urban Economics and Economic Development Boca Raton, FL: Auerbach Publications unicipal Land Policies and Their Outcomes ed. G. K. Ingram and Y. H. Hong. Cambridge, MA: Lincoln Institute of Land Policy. Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. England. Cambridge, MA: Lincoln Institute of Land Policy, May: 27 47 Gaffney, Mason. (2008). "Keeping land in capital theory: Ricardo, Faustmann, Wicksell, and George", The American Journal of Economics and Sociology, Vol. 67 No. 1, pp. 119 41. George, Henry. ([1879] 1938). Progress and Poverty New York: Modern Library. Grosskopf, S. ( 1981 ) of a general equilibrium model with recent empirica l estimates of several key American Journal of Economics and Sociology 40 (2): 207 215. Haughwout, A. F. ( 2004 ) City taxes, city spending: Essays in honor of Dick Netzer ed A. E. Schwartz. Northampt on, MA: Edward Elgar. Huhne, C. ( 2004 ) Why we should follow Pittsburgh New Statesman (London, England: 1996) 133 38 9. Retrieved January 19, 2010, from OmniFile Full Text Mega database. Kumar, R. (2005). Research Methodolo gy: A Step by Step Guide for Beginners. Second Edition. Thousand Oaks, CA: Sage Publications. Lusht, K. ( 1992 ) The site value tax and residential development. Working paper, Lincoln Institute of Land Policy, Cambridge, MA.

PAGE 151

151 Mathis, E. J. & C. E. Zech. ( 198 2 ) 5. Mathis, E. J. & C. E. Zech. ( 1983 ) 48. McDonough, C. C., & Si hag, B. S. (1991). The incorporation of multiple bases into shift share analysis. Growth & Change 22(1), 1. Retrieved from EBSCO host. Mills, D. E. ( 1981a ) National Tax Journal 34 (1):125 129. Mills, D. E. ( 19 81b ) Regional Science and Urban Economics 11(2): 239 254. Mills, D. E. ( 1983 ) Research in urban economics ed. J. V. Henderson. London: JAI Press. Mills, E. S. Land value taxation: Can it and will it work today? Cambridge, Mass: Lincoln Institute of Land Policy. Nechyba, T. J. ( 1998 ) Land Value Taxation: Can it and will it work today? Ed. D. Netzer. Cambridge, MA: Lincoln Institute of Land Policy. Netzer, D. ( 1998a ) Land Value Taxation: Can it and will i t work today? Cambridge, MA: Lincoln Institute of Land Policy. Netzer, D. ( 1998 b ) Land Value Taxation: Can it and will it work today ? Ed. D. Netzer. Cambridge, MA: Lincoln Institute of Land Policy Netzer, D. ( 1998c ) feasibility of land value taxation in rich Land Value Taxation: Can it and will it work today? Ed. D. Netzer. Cambridge, MA: Lincoln Institute of Land Policy. Oates, W. E. & R. M. Schwab. ( 1997 ) Pitt National Tax Journal 50 (March): 1 21. Oates, W. E. & Schwab, R. M. ( 2009 ) In Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. England. Cambridge, MA: Lincoln Institute of Land Policy : 51 71. Plassmann, F. & T. N. Tideman, ( 2000 ) effect of two Journal of Urban Economics 47 (2):216 247.

PAGE 152

152 Plummer, E. ( 2009 ) Land Value Taxation, Theory, Evidence and Practice Eds. Richard F. Dye and Richard W. England. Cambridge, MA: Linco ln Institute of Land Policy : 73 98. Pollakowski, H. O. ( 1982 ) Working paper, Lincoln Institute of Land Policy, Cambridge, MA. Reeb, D. J. ( 1998 ) The adoption and repeal of the two rate property tax in Amsterdam, New York. Working paper, Lincoln Institute of Land Policy, Cambridge, Massachusetts Schwab, R. M. & A. R. Harris. ( 1997 ) Taxing simply, taxing fairly Washington, DC: District of Columbia Tax Revision Commission. Stevens, S. K. (1955). "A Century of Industry in Pennsylvania," Pennsylvania History, 22 (1) 49 68. Retrieved 20 Sept 2011 from http://dpubs.libraries.psu.edu/DPubS?service=Repository&version= 1.0&verb=Diss eminate&handle=psu.ph/1141401058&view=body&content type=pdf_1# Tideman, T. N. ( 1982 ) 109 111. Tideman, T. N. ( 1998 ) Environm ental Protection, Congestion, Efficient Resource Use, Population and Land Value Taxation: Can it and will it work today? Ed. D. Netzer. Cambridge, MA: Lincoln Institute of Land Policy Tideman, T. N. & C. Johnson, ( 1995 ) A statistical analysis of graded property taxes in Pennsylvania Cambridge, MA: Lincoln Institute of Land Policy. Tiits, T. ( 2008 ) Making the property tax work: Experiences in developing and transitional countries Eds. R. W. Bahl, J. Marti nez Vasquez, and J. M. Youngman. Cambridge, MA. Lincoln Institute of Land Policy. Turnbull, G. K. ( 1988 ) Journal of Real Estate Finance and Economics 1, 393S403. U S Census Bureau. (Various years ). U S Census of Population and Housing. Retrieved from http://factfinder.census.gov/servlet/DatasetMainPageServlet?_program=DEC&_ss ubmenu I=datasets_0&_lang=en U.S. Census Bureau. (2004). Definition: Ratio of income to poverty. Retrieved from: https://ask.census.gov/app/answers/detail /a_id/319/~/definition%3A ratio of income to poverty

PAGE 153

153 U S Census Bureau. (2009). 2005 2009 American Community Survey. Retrieved from http://factfinder.census.gov/servlet/DatasetMainPageServlet?_program=ACS&_su bmenuId=datasets_1&_lang=en&_ts= U S Census Bureau (2010) 2010 U S Census of Population and Housing Retrieved from http://2 010.census.gov/2010census/ U S Department of Labor, Bureau of Labor Statistics. (2011). Retrieved from http://data.bls.gov/LOCATION_QUOTIENT/servlet/lqc.ControllerServlet Land Value Taxation: The Equitable and Efficient Source of Public Finance ed. Kenneth C. Wenzer. Armonk, NY: M. E. Sharpe.

PAGE 154

154 BIOGRAPHICAL SKETCH Robert J. Murphy, Jr. was born in Albany, New York in 1986 and raised in Mechanicville, New York. He would go on to graduate from Mechanicville High School in 2005 where he was salutatorian of his senior class. After high school, Robert would go on to attend the Rensselaer Po lytechnic Institute in Troy, NY. While at Rensselaer, he would spend a fall semester studying at the Universit Comerciale Luigi Bocconi in Milan, Italy. In May of 2009, Robert would earn a Bachelor of Science degree from the Lally School of Management a nd Technology with a minor in economics. Toward the end of his tenure at Rensselaer, Robert would garner interest in city planning particularly in his hometown. He channeled this interest by deciding to pursue a Master of Arts in Urban and Regional Pla nning at the Universi ty of Florida in Gainesville, Florida concentrating his studies in economic development planning. and preserving the uniqueness and economic viabi lity of cities, towns, and regions.