PB-2006-1 | February 2006

Regulation and the Rise of Housing Prices in Greater Boston By Edward L. Glaeser, Jenny Schuetz and Bryce Ward, Harvard University

In well-functioning markets, when Supply, however, is not keeping up: Edward L. Glaeser prices rise, supply increases, and then • In the 1960s, there were 172,459 Edward L. Glaeser is the Fred and Eleanor prices stop rising and sometimes even units permitted in the Boston Glimp Professor of Economics in Harvard’s fall. By this defi nition, the housing Faculty of Arts and Sciences and is director metropolitan area; in the 1980s, of Harvard’s Rappaport Institute for Greater market in the greater Boston area is 141,347. However, despite the Boston. He teaches urban and social not working. economics and has published papers on sharp rise in prices in the 1990s, cities, economic growth, and housing The market is sending clear signals only 84,105 units were permitted prices. about the demand: in that decade. Jenny Schuetz • Between 1980 and 2004, housing • The decline in permits has been Jenny Schuetz is a doctoral student in public policy at Harvard’s Kennedy School prices in three of the Census particularly striking for units in of Government. Her primary research Bureau’s divisions of the Boston multi-family buildings. In the interests include land-use regulation, housing policy and fi nance. metropolitan area grew by 1960s, less than 50 percent of all between 179 and 210 percent permits in the Boston metropolitan Bryce Ward (adjusted for infl ation), which area were for single-family homes. Bryce Ward is a doctoral student in Harvard’s Department of Economics. His made these areas—Boston- In the 1990s, over 80 percent of research interests are the eff ects of zoning Quincy, Cambridge, Newton, and all permits were for single-family on housing and the eff ects of distance/ space on social participation. Essex County—second through homes. This policy brief is based on “Regulation fourth in the nation behind only Some of the price increase can be and the Rise of Housing Prices,” a working the New York area’s Nassau- paper by Glaeser, Schuetz and Ward that explained by the region’s dramatic was released in January by the Rappaport Suffolk Division. economic renaissance in the past Institute and the Pioneer Institute. The full paper is available online at http://www. • According to data from the National three decades. However, other regions ksg.harvard.edu/rappaport/downloads/ Association of Realtors in the third have boomed without experiencing housing_regulations/ regulation_ housingprices.pdf quarter of 2005, the median sales dramatic increases in house prices. price for existing single-family Until the last quarter, for example, Rappaport Institute for Greater Boston homes in the Boston metropolitan median housing prices in the Phoenix Harvard University area was $430,900, more than any area were less than $200,000 and John F. Kennedy School of Government other region in the continental in the Houston area the current 79 JFK Street, Cambridge, MA 02138 Telephone: (617) 495-5091 United States except for portions median sales price is $142,000. Local http://www.ksg.harvard.edu/rappaport of California, greater New York governments in Phoenix area handed City, and the Washington, D.C. out 57,273 permits for single-family Pioneer Institute for Public Policy Research metropolitan area. homes during 2004, Las Vegas area 85 Devonshire Street, Boston, MA 02109 localities 35,579—local governments Telephone: (617) 723-2277 in the greater Boston area only 5,001. http://www.pioneerinstitute.org Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

Is Greater Boston Running Out of Land? and the Rappaport Institute developed a unique new dataset on land-use regulation in 187 cities There are two theories about why so little new and towns in eastern and central Massachusetts. housing is being built in Greater Boston. It Working under the direction of Pioneer’s Amy may simply be that the area has run out of land. Dain, researchers answered more than 100 After all, the Boston metropolitan area is one questions about each community’s land-use of the country’s most dense metropolitan areas. regulations by reviewing offi cial documents Alternatively, the shortfalls in supply may be and interviewing local offi cials, who were the result of restrictive land use regulations. subsequently given the opportunity to review There is little evidence to support the view that the data about their community. greater Boston simply lacks the land to build The most striking fact that emerges from the new homes. Within the urban core, it would data is that developers face an incredibly be quite feasible technically to build taller heterogeneous set of local regulatory regimes. buildings. In fact, with strong support from the This heterogeneity begins with minimum lot city, a host of new high-rise residential housing size, which remains the most important restraint has been built or is being built in the heart of on the use of land. The 22 municipalities in the Boston. While densities outside of the core are region with average minimum lot sizes of less high relative to the United States as a whole, than a quarter of an acre contain more than 25 they are still quite low, averaging 1.4 acres percent of the region’s population. In contrast, per home for communities within 50 miles of the 14 municipalities where minimum lot size Boston. Moreover, if land were just scarce, is greater than 70,000 square feet (1.625 acres) then the price of a quarter acre of land would cover ten percent of the region’s land but hold be the same whether it extends an existing lot only four percent of its population. or if it sits under a new home. Communities have at their disposal a number of Data suggests that regulation, other regulations that they can use to limit new not density, has caused low construction. levels of new construction and • Growth caps and phasing schedules. high housing prices in Greater Communities can use growth caps to limit Boston. the number of new units that can be built during a given year, or phasing schedules to limit the number of units per year that However, Glaeser and Gyourko (2002) fi nd can be built within a single subdivision. We that a quarter-acre is worth 20 times more in identifi ed 54 communities that made use of greater Boston if it sits under a new house growth limitations, the vast majority having than if it extends the lot of an existing house, adopted them in the last ten years. suggesting that surviving the regulatory • The prohibition of irregularly shaped lots. process adds enormous value. More than half of the communities with the largest minimum lot sizes make it even Greater Boston’s Regulatory Web harder to meet their standards by requiring Such data suggest that regulation, not density, that those lots be suffi ciently compact. has caused low levels of new construction • Wetland regulations. and high housing prices in Greater Boston. To More two-thirds of the 187 communities help test this claim, over the past two years the have wetlands bylaws or ordinances that are Pioneer Institute for Public Policy Research 2 Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

Total Housing Permits Issued, Boston Area

Source: US Census Bureau. Permits are for Boston Metropolitan Statistical Area

stricter than state wetland regulations. Only minimum lot sizes, cluster zoning typically a handful of these bylaws or ordinances allows almost a two-thirds reduction in the were adopted before 1980. More than 50 minimum lot-size requirement for each communities adopted them in the 1980s, home. At the same time, however, many and more than 50 have adopted them since of the communities allow no more units in 1990. a development built under cluster zoning • Septic-system regulations. than would have been allowed under a We counted 109 communities with conventional zoning plan. septic-system regulations stricter than the • Inclusionary zoning provisions often allow state’s standards, which is two-thirds of builders to construct at higher densities municipalities that are not entirely served if they include some housing units by public sewer systems. designated as affordable to lower moderate- • Subdivision rules. income households. Ninety-nine of the All but six communities have rules for municipalities in our sample have adopted subdivisions. Some adopted the regulations some type of inclusionary zoning provision before 1950 and most did so by 1980. and nearly half of those have adopted the More than 70 amended their bylaws after provision since 2000. Seventy-seven of 2000. these communities offer a density bonus for including the affordable units. However, While communities in the greater Boston the provisions have never been used in at area have also adopted measures that relax least 43 of the 99 communities. minimum lot-size requirements, they often fi nd ways to discourage their use. • Age-restricted zoning is often used to allow smaller minimum lot sizes if the • Cluster provisions allow developers to development is open only to older adults. build at higher densities if they set aside Almost 60 percent of those communities some amount of open space. The lot size with more than 20,000 square foot reductions due to cluster zoning are quite minimum lot sizes have some form of dramatic. In communities with large provision for such age-restricted housing. 3 Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

More than 40 percent of those communities twelve percent drop in new construction. The with minimum lot sizes that are greater estimated effects of each form of regulation than 35,000 square feet have provisions individually are inconclusive; the magnitude that allow for age-restricted multi-family of the effects is imprecisely estimated; and we housing. cannot be certain that the effects are statistically different from zero. However, combining all Impacts on Permits three forms of regulations into one index, we obtain statistically robust results indicating that While these regulations are striking and di- each additional form of regulation is associated verse, it is less clear which of them actually with a ten percent decline in annual permits. matter. To answer this question, we connected The degree of correlation across these three the new dataset with data on permits by local- kinds of regulation, however, prevent us from ity going back to 1980, Banker and Tradesmen coming to a reliable conclusion about the data on house sales in the greater Boston area, degree to which each individually contributes U.S. Census data going back to 1910, and the to the shortfall in new construction permits. Massachusetts GIS system, which collected data on minimum lot size requirements across The reductions in permits caused the state in 1999–2000. by the regulations has had a The evidence linking minimum lot size to signifi cant eff ect on regional development is persuasive. On average, as housing prices. average minimum lot size increases by one- quarter of an acre, there were approximately What about features that alleviate the burdens ten percent fewer houses in 1970, nine percent of zoning? Adoption of cluster zoning is fewer houses in 2000, and ten percent fewer correlated with an increase in the amount of houses permitted between 1980 and 2002. new development, but we were not able to These results are, perhaps, unsurprising, but discern an impact of inclusionary zoning. they do confi rm the important role that zoning We also were unable to assess the impact of has on new development. Chapter 40B, the Massachusetts Anti-Snob Perhaps, more surprisingly, the connection Zoning Act, which allows the state to overrule between minimum lot size and development local land-use decisions for projects, because is declining over time, as even places with it impacts most municipalities. Nonetheless, smaller minimum lot sizes radically reduce the we can say that the more than 30,000 units amount of new construction they allow. constructed under Chapter 40B have accounted It is less clear which of three broad areas of for a signifi cant part of new development in additional regulation have had the greatest many areas. impact on new construction: wetland Housing Prices regulations, septic rules, or subdivision policies. We found that when localities impose The reduction in permits caused by the wetlands regulations stricter than those regulations has had a signifi cant effect on imposed by the state, new construction appears regional housing prices. Since 1990, for to fall by about ten percent. When localities example, the housing stock in greater Boston impose rules for septic systems that are stricter increased by only nine percent. Published than state standards, new construction falls by estimates of housing demand elasticities about four percent. Adoption of subdivision (Ermisch, Findlay, and Gibb 1996), suggest rules, fi nally, is associated with about a that if the housing stock had instead increased 4 Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

Distribution of Minimum Lot Size Requirements

Source: Massachusetts GIS Database

by 27 percent, as it did from 1960 to 1975, 1995 suggests that more restrictive land-use housing prices would be 23 to 36 percent regulations are more potent at high points of the lower. That is, the median house price, real estate cycle. which is now $431,900, would have been Median sales price, however, does not control as low as $276,100. for differences in housing characteristics or While we can show that regulations reduce the land area under the median home that new construction permits, and connect that is being sold. Housing units in areas with lack of supply to high housing prices in larger lots may be more expensive because the greater Boston area, it is more diffi cult they are larger and have more land. When we to estimate the price of housing directly control for housing characteristics such as the from the degree of regulation. Housing number of rooms, age of the home, internal markets are regional not local: more square footage, and total acreage of the lot, restrictions in Wellesley will not only raise one additional acre in the minimum lot size prices in Wellesley, but also in neighboring is associated with between an 11.5 and 13.8 Needham, even if Needham has less percent increase in housing prices, depending stringent land-use restrictions. Because on what other factors we control for. This is we lack the kind of clear comparisons we less than the effect when we do not take into had with the effect on permits, we have to account actual acreage under the housing unit, make do with less precision and certainty. but is still signifi cant. Nevertheless, land-use regulations do Land-use regulation has also reduced the seem to have an impact on locality-specifi c amount of affordable housing in the greater prices: an additional acre in minimum Boston area. To assess this impact, we lot size raised the median sales prices of calculated the share of sales in the community homes in the locality in question by 15.8 where an average resident of the region could percent in 1987, 11.3 percent in 1995, pay for the interest on the purchase price of and 19.5 percent in 2001. That impacts the home with 30 percent of his or her income. were higher in 1987 and 2001 than in We fi nd that as minimum lot size increases by 5 Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

one acre, the share of homes that qualify as bulk of its local aid to successfully encourage affordable by this defi nition drops by 8-to-20 new construction. percent. 2. The state could follow, more intrusively, the lead of Chapter 40B and give state Addressing the Problems or regional entities the power to overrule The current system has four structural features local land-use decisions in communities that must be addressed if proposals for change with low density levels, high prices, and are likely to be effective. few permits. Such an override, moreover, 1. While individual communities have every should be linked with impact fees at a level incentive to impede new construction with set by the state. Overriding local control is land-use regulations, those who have yet to sure to be unpopular, but it is also the surest buy and those who reside in surrounding way of breaking local bottlenecks on new communities, whether individuals or construction. businesses, suffer. The current system has four 2. Localities have demonstrated remarkable structural features that must resilience and creativity in keeping the supply of housing low. If the state tries be addressed if proposals for to limit the restrictions along any one change are likely to be eff ective. dimension, communities will increase them on one or more other dimensions. 3. The state could take policy actions that 3. The high degree of ambiguity in regulations clarify rights and limit the potential for and uncertainty in the permitting process litigation while simultaneously increasing increases costs for developers and protections for current homeowners, thus encourages frivolous court challenges. It is improving the lot of both developers both hard and expensive for developers to and local homeowners. Such measures raise money, and diffi cult for developers, might include requiring plaintiffs who local offi cials, and abutters to negotiate unsuccessfully challenge a project in court binding agreements. to pay a fee or to pay the developers’ legal costs. 4. With only limited procedures for allowing developers to compensate current residents 4. The state could substitute existing regulations and communities for the negative impact with a well-designed impact-fee system of new development, the current system is that would reduce uncertainty, promote new economically ineffi cient. construction, and enhance the welfare of developers, abutters, and communities. We Four policy approaches could address the want to emphasize, however, that impact fees problems created by these features. could only increase new construction and 1. The state could alter local incentives by housing affordability if they replace existing using state aid to reward localities that barriers to new construction. encourage new construction and punish If the residents and businesses in greater Boston those who discourage it. While the recently are seriously interested in making affordable passed Chapters 40R and 40S, which are housing a reality, they must lower the barriers designed to eliminate fi scal problems created against new construction. However, since by new development, are small steps in the there is no reason to expect that localities will right direction, the state needs to use the act against the self-interest of homeowners, 6 Regulation and the Rise of Housing Prices Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

it is up to the state to take action to change ABOUT THIS POLICY BRIEF the schedule of incentives and relieve the This policy brief is part of the Initiative on Local externalities burdening those who have yet to Housing Regulation, a joint effort of the Pioneer buy a home and businesses. Because the only Institute for Public Policy Research and Harvard University’s Rappaport Institute for Greater Boston. way to reduce the price of something is to As part of this initiative, researchers at the Pioneer produce more of it, it is logically incoherent to Institute and the Rappaport Institute have assembled be both an advocate of affordable housing and and coded a database on zoning codes, subdivision an opponent of new construction. requirements, and environmental regulations that as of 2004 governed land use in 187 communities in REFERENCES eastern and central Massachusetts. The searchable “The Impact of Zoning on Housing database is available at http://www.pioneerinstitute. Affordability,” org/municipalregs/. The site also houses summary by Edward L. Glaeser and Joseph Gyourko (HIER reports, analyses of the data, and a downloadable Discussion Paper Number 148 (2002)). version of the database in formats that can be used for econometric analyses. In coming months, the “The Price Elasticity of Housing Demand in Pioneer Institute and the Rappaport Institute will Britain: Issues of Sample Selection,” also be issuing papers and policy briefs, some by John F. Ermisch, Jeanette Findlay, and Kenneth jointly and some individually, on land-use regulation Gibb, Journal of Housing Economics, 5(1): 64-86. in greater Boston.

INSTITUTIONS

The Pioneer Institute for Public Policy Research, founded in 1988, is an independent, non- profi t, public policy research insititute based in Boston that generates and markets new and practical public policy ideas and peer-reviewed scholarship. Pioneer Institute research explores the application of market principles to state and local policy to advance the core values of an open society - individual freedom and responsibility, economic opportunity, social mobility, and limited government.

The Rappaport Institute for Greater Boston at Harvard University strives to improve the region’s governance by attracting young people to serve the region, working with scholars to produce new ideas about important issues, and stimulating informed discussions that bring together scholars, policymakers, and civic leaders. The Rappaport Institute was founded and funded by the Jerome Lyle Rappaport Charitable Foundation, which promotes emerging leaders in Greater Boston. RAPPAPORT Institute for Greater Boston Kennedy School of Government, Harvard University John F. Kennedy School of Government 85 Devonshire Street Harvard University Boston, MA 02109 79 John F. Kennedy Street 617-723-2277 Cambridge, MA 02138 http://www.pioneerinstitute.org 617-495-5091 [email protected] [email protected] http://www.rappaportinstitute.org

7 Rappaport Institute | PioneerTaubman Institute Center POLICY BRIEFS

PREVIOUS RAPPAPORT INSTITUTE PB-2006-1, January 2006 POLICY BRIEFS “Regulation and the Rise of Housing Prices in Greater Boston” PB-2004-1, October 2004 by Edward L. Glaeser, Jenny Schuetz and Bryce “Can Social Capital Last: Lessons from Boston’s Ward (Harvard University) Villa Victoria Housing Complex,” by Mario Luis Small (Princeton University) UPCOMING EVENTS

PB-2005-1, January 2005 Developing Concerns: Lessons from Arlington, “Betting the Future: The Economic Impact of Massachusetts’ Changing Views on Development Legalized Gambling,” Tuesday, February 28 at 12:00 p.m. by Phineas Baxandall (Rappaport Institute for Taubman AB, 5th fl oor Taubman Building Greater Boston) and Bruce Sacerdote (Dartmouth 15 Eliot Street, next to the Charles Hotel College) Crime, Time and Employment: Realistic Options PB-2005-2, February 2005 for Helping People Leaving Prison Find Better “Needed Corrections: Promising Strategies for Jobs Improving Massachusetts’ Prisons and Jails,” Monday, March 13 at 5:00 p.m. by Anne Morrison Piehl (Kennedy School of Bell Hall, 5th fl oor of the Belfer Building Government) At the corner of John F. Kennedy and Eliot Streets

PB-2005-3, March 2005 Guarding the Town Walls: Mechanisms and “Standards-Based Education Reform in the Motives for Redistricting Multifamily Housing in Computer Age: Lessons from Boston’s Murphy Massachusetts School,” Wednesday, March 22 at 12:00 p.m. by Frank Levy (Massachusetts Institute of Bell Hall, 5th fl oor of the Belfer Building Technology) and Richard Murnane (Graduate At the corner of John F. Kennedy and Eliot Streets School of Education, Harvard University) Extending Boston’s Four-Century History of PB-2005-4, April 2005 Innovation “Smart Growth: Education, Skilled Workers, Thursday, April 6 at 12:00 p.m. and the Future of Cold-Weather Cities,” Taubman AB, 5th fl oor of the Taubman Building by Edward L. Glaeser (Harvard University) 15 Eliot Street, next to the Charles Hotel

PB-2005-5, September 2005 Police Oversight in the Face of Terrorism: “Creating an Effective Foundation to Prevent Lessons from London Youth Violence: Lessons Learned from Boston in Tuesday, April 11 at 6:00 p.m. the 1990s,” John F. Kennedy, Jr. Forum, 79 John F. Kennedy St. by Anthony A. Braga (Kennedy School of Government) and Christopher Winship (Faculty Planning at the Cutting Edge: How are the of Arts and Sciences and Kennedy School of Communities in the I-495 Corridor Responding Government, Harvard University) to Rapid Growth? Wednesday, April 12 at 12:00 p.m. PB-2005-6, October 2005 Taubman AB, 5th fl oor Taubman Building “Crowd Control That Can Kill: Can American 15 Eliot Street, next to the Charles Hotel Police Get a Grip on Their New, ‘Less Lethal’ Weapons Before They Kill Again?” The Boston Youth Violence Project: Current by Christopher Stone (Kennedy School of Trends and Future Challenges Government), Brian Buchner and Scott Dash Monday, April 17 at 5:00 p.m. (Police Assessment Resource Center) Bell Hall, 5th fl oor of the Belfer Building At the corner of John F. Kennedy and Eliot Streets PB-2005-7, November 2005 “Local Services, Local Aid and Common Challenges” by Phineas Baxandall (Rappaport Institute for Greater Boston) 8