Skip to main content

Effects of Rent Control on Housing Dynamics (public)

Research reportView only

Create alertChat

September 12, 2025

What are the effects of rent control policies on housing market dynamics and tenant stability?

Rent control policies yield mixed results, decreasing rents in controlled units by 4-47% and extending tenant stays by up to six years, while simultaneously reducing housing supply and increasing eviction filings, with effects varying by policy design and local market conditions.

Abstract

Rent control policies yield mixed effects on housing market dynamics and tenant stability. In regulated markets, studies report rent declines ranging from 4% to 47% in controlled units while some analyses note rent increases in uncontrolled or post‐decontrol sectors. In one San Francisco study, rents rose concurrently with a 15% reduction in housing supply and a 20% drop in tenant mobility. Other research documents tenants staying an extra six years under tighter regulation, although reports of 83–127% higher eviction filings suggest displacement risks in certain settings.

Separate analyses show shifts in market composition. For example, one study cited annual rent subsidies of about US$2440 (roughly 27% of income), and another observed reduced racial disparities accompanied by increased minority presence. These findings, based on quasi‐experimental and econometric approaches across diverse jurisdictions, indicate that the impacts of rent control vary with policy design and local market conditions.

Methods

We analyzed 40 sources from an initial pool of 989, using 8 screening criteria. Each paper was reviewed for 5 key aspects that mattered most to the research question. More on methods

Papers identified with Elicit search

n = 989

Papers screened using: Intervention Type, Outcome Measures, Study Design, Geographic Context, Public/Social Housing Focus, Housing Voucher/Assistance Focus, Empirical Data, Comparison Groups

n = 989

Papers screened out

n = 949

Papers included for extraction

n = 40

Press enter or space to select a node.You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel.

Press enter or space to select an edge. You can then press delete to remove it or escape to cancel.

Paper search

Using your research question “What are the effects of rent control policies on housing market dynamics and tenant stability?”, we searched across over 126 million academic papers from the Semantic Scholar corpus. We retrieved the 989 papers most relevant to the query.

Screening

We screened in sources based on their abstracts that met these criteria:

• Intervention Type: Does this study examine rent control, rent stabilization, or rent regulation policies?
• Outcome Measures: Does this study measure at least one of the following: housing market outcomes (rental supply, housing prices, construction rates, property maintenance, market competition) OR tenant stability outcomes (residential mobility, displacement rates, tenure length, eviction rates, housing security)?
• Study Design: Is this study one of the following: randomized controlled trial, quasi-experimental design, natural experiment, longitudinal study, cross-sectional study with comparison groups, systematic review, or meta-analysis?
• Geographic Context: Was this study conducted in urban or metropolitan areas with rental housing markets?
• Public/Social Housing Focus: Is this study NOT focused solely on public housing or social housing programs without rent control components?
• Housing Voucher/Assistance Focus: Is this study NOT examining only housing voucher or rental assistance programs without rent regulation?
• Empirical Data: Is this study NOT purely theoretical or modeling without empirical data?
• Comparison Groups: Is this study NOT a case study or descriptive study without comparison groups or control conditions?

We considered all screening questions together and made a holistic judgement about whether to screen in each paper.

Data extraction

We asked a large language model to extract each data column below from each paper. We gave the model the extraction instructions shown below for each column.

• Study Design Type:

Identify the primary research design used in the study. Look in the methods section for explicit description of the study design.

Possible design types include:

• Difference-in-differences
• Quasi-experimental
• Empirical analysis
• Econometric study
• Cross-sectional analysis

If multiple design elements are used (e.g., difference-in-differences with discontinuity design), list all relevant approaches. Be precise about the specific methodological approach.

If the design type is not clearly stated, review the analytical methods and statistical techniques used to infer the most appropriate design classification.

• Geographic Setting and Context:

Extract the specific geographic location(s) where the study was conducted. Include:

• Country
• City/metropolitan area
• Specific neighborhoods or regions (if applicable)

If multiple geographic areas are studied, list all of them. Note any specific contextual details about the housing market that are relevant to understanding the rent control policy (e.g., housing density, urban vs. suburban, economic conditions).

If geographic context is not explicitly stated, check methods, data sources, and results sections for implicit geographic information.

• Rent Control Policy Details:

Extract comprehensive details about the rent control policy:

• Year of policy implementation
• Specific type of rent control (first-generation, second-generation, etc.)
• Key policy mechanisms (e.g., price caps, exemptions)
• Scope of application (which housing units are covered)

Look in introduction, methods, and policy description sections. If multiple policy aspects are discussed, provide a comprehensive summary.

If policy details are complex, create a bulleted list to capture nuanced information. Note any explicit exceptions or special conditions in the policy.

• Policy Impact Outcomes:

Identify and extract the specific outcomes measured related to rent control policy:

• Rental prices (regulated and unregulated sectors)
• Housing supply
• Land values
• Maintenance investments
• Tenant stability
• Demolition rates
• Housing market dynamics

Extract quantitative values where available. If multiple outcomes are measured, list all with their specific measurements and statistical significance.

Prioritize outcomes directly addressing rental prices, housing supply, and market dynamics. Look in results and discussion sections for comprehensive outcome reporting.

• Key Quantitative Findings:

Extract the primary quantitative results, focusing on:

• Magnitude of effects
• Statistical significance
• Confidence intervals or standard errors (if provided)
• Directional impact of rent control policy

Prioritize results that directly answer the research question about housing market dynamics and tenant stability.

If multiple statistical models or analyses are presented, extract results from the primary or most comprehensive analysis. Include specific numeric values and percentage changes where available.

Ensure extracted findings are directly quoted or paraphrased from the results section, maintaining the original statistical context.

Results

Characteristics of Included Studies

Study

Study Location/Period

Rent Control Policy Type

Primary Outcomes Measured

Methodology

Full text retrieved

Diamond et al., 2019a

San Francisco, USA; 1994 onward

Not explicitly labeled; limits rent increases, multi-family 2-4 units built 1900-1990

Rental prices, housing supply, tenant stability, maintenance

Quasi-experimental

Yes

Jofre-Monseny et al., 2023

Catalonia, Spain; 2020

Not specified; price caps, high coverage

Rental prices, housing supply

Difference-in-differences, event-study, quasi-experimental

No

Munch & Svarer, “Rent Control and Tenancy Duration”

Denmark

Not specified in abstract

Tenancy duration, mobility

Econometric study

No

Mense et al., 2017

Germany; 2015

Second-generation; exception for new units

Rental prices, land values, supply, maintenance

Difference-in-differences with discontinuity, quasi-experimental

No

Breidenbach et al., 2021

Germany; 2015

Not specified in abstract

Rental prices, quality

Quasi-experimental (triple difference event study)

No

Mense et al., 2019

Germany; 2015

Second-generation; rent ceiling, multiple exceptions

Rental prices, supply, land values, maintenance, mobility

Quasi-experimental (difference-in-differences with discontinuity)

Yes

Gyourko & Linneman, 1989

New York City, USA; 1943 onward

First-generation; rents frozen, later exemptions

Rental prices, supply, tenant stability, maintenance

Econometric, cross-sectional

Yes

Autor et al., 2012

Cambridge, MA, USA; 1970-1995

First-generation; strict caps, no vacancy decontrol

Rental prices, supply, land values, maintenance, turnover

Quasi-experimental, difference-in-differences, econometric

Yes

Early & Phelps, 1999

USA; 1984-1996

Not specified; new construction exempt

Rental prices (uncontrolled sector), supply

Econometric study

No

Fallis & Smith, 1985

Toronto, Canada

Not specified; exemptions

Rental prices (controlled/uncontrolled)

Econometric, cross-sectional

No

toof

Pageof

View 30 more rows

Across 40 studies of rent control policies, the following patterns were observed:

Primary Outcomes Measured:

• Rental prices:Measured in 25 studies.
• Housing supply (including new construction and conversions):Measured in 20 studies.
• Tenant stability (including tenure duration, mobility, turnover, and displacement):Measured in 16 studies.
• Maintenance, quality, investment, or renovations:Measured in 9 studies.
• Evictions or eviction filings:Measured in 3 studies.
• Land values, property values, or wealth transfer:Measured in 5 studies.
• Vacancy rates:Measured in 2 studies.
• Market dynamics (including spillovers, misallocation, search/matching costs, and disequilibrium):Measured in 6 studies.
• Rent burden, rent discounts, or rent subsidy:Measured in 4 studies.
• Community composition or segregation:Measured in 2 studies.
• Owner/tenant knowledge, queue time, and income distribution:Each measured in 1 study.
• Other outcomes (such as distribution, race, tenant welfare, and consumption):Found in 5 studies.

Rent Control Policy Type:

• First-generation rent control:Studied in 7 studies.
• Second-generation rent control:Studied in 10 studies.
• Third-generation rent control:Studied in 1 study.
• Rent stabilization:Explicitly labeled in 3 studies.
• No clear specification of policy type:21 studies.
• User value system (other):1 study.

Methodology:

• Quasi-experimental methods (including difference-in-differences, event study, regression discontinuity, interrupted time series, and triple difference):Used in 24 studies.
• Econometric methods (including cross-sectional and empirical econometric analysis):Used in 21 studies.
• Empirical analysis (not otherwise specified):Used in 11 studies.

We did not find mention of some outcomes or policy types for several studies, either due to lack of reporting or lack of access to the full text.

Effects

Effects on Housing Market Dynamics

Study

Outcome Measure

Effect Direction

Effect Size

Statistical Significance

Diamond et al., 2019a

Market rents, supply

Increase in rents, decrease in supply

15% supply reduction, 20% decrease in mobility

No mention found

Jofre-Monseny et al., 2023

Average rents

Decrease in rents

4–6% reduction

No mention found

Munch & Svarer

Tenancy duration

Increase in duration

+6 years (most vs. least regulated)

No mention found

Mense et al., 2017

Rents (regulated/unregulated), land values

Increase in unregulated, decrease in regulated, increase in land values

No mention found

No mention found

Breidenbach et al., 2021

Rental prices

Decrease in rents (short-term)

Up to 5% (all), 9% (some types)

Significant, effect vanishes after 1 year

Mense et al., 2019

Regulated/free-market rents

Decrease in regulated, increase in free-market

No mention found

No mention found

Gyourko & Linneman, 1989

Rent subsidy

Increase in subsidy

$2440/year (27.2% income)

95% confidence interval: $2367–$2511

Autor et al., 2012

Rents, property values

Increase in rents post-decontrol

40% (controlled), 13% (never-controlled)

No mention found

Early & Phelps, 1999

Uncontrolled rents

Increase in rents

Positive, significant

Yes

Fallis & Smith, 1985

Rent difference

Attributable to controls

~50% of difference

No mention found

toof

Pageof

View 30 more rows

Across 40 studies, the following patterns were observed in effect direction and statistical significance:

• Decrease in rents:Reported in 11 studies.
• Increase in rents:Reported in 10 studies. Some studies reported both increases and decreases, depending on market segment (for example, regulated versus unregulated units).
• Decrease in supply or construction:Found in 4 studies.
• Increase in supply or construction:Found in 3 studies.
• Increase in tenant stability or tenure:Found in 2 studies.
• Decrease in tenant stability or tenure:Found in 1 study.
• Increase in evictions or displacement:Found in 3 studies.
• Decrease in evictions or displacement:Found in 1 study.
• Increase in property or land values:Found in 1 study.
• Decrease in property or land values:Found in 1 study.
• Increase in vacancy:Found in 2 studies. We did not find studies reporting a decrease in vacancy.
• Increase in rent burden, savings, discounts, or subsidy:Found in 5 studies.
• Mixed, minimal, or no effect:Reported in 5 studies.
• Other outcomes (queue times, demographic changes, misallocation costs):Reported in 2–3 studies.

Statistical significance:

• Explicit statements of statistical significance:Found in 9 studies, based on available full texts or abstracts.
• Explicit statements of non-significance:Found in 2 studies.
• No mention of statistical significance:In the available full texts or abstracts for the remaining 29 studies.

The most common outcome measures were rents or rental prices, supply or construction, tenant stability, tenure or mobility, evictions or displacement, property or land values, and rent burden, savings, or subsidy. We did not find mention of outcome measure information in the available full texts or abstracts for some studies, and some studies reported multiple outcomes.

Effects on Tenant Stability and Demographics

Tenant Composition and Income Distribution

Study

Outcome Measure

Effect Direction

Effect Size

Statistical Significance

Gyourko & Linneman, 1989

Rent subsidy, tenure

Increase in subsidy, increase in tenure

$2440/year, significant tenure increase

95% confidence interval: $2367–$2511

Donner & Kopsch, 2023

Subsidy, queue, income

Increase in subsidy for high-income, increase in queue

21 vs. 10 years (high vs. low subsidy)

No mention found

Chen et al., 2023

Rent discount, race

Decrease in racial disparity, increase in discount

$410/month, $4–5.4B/year

No mention found

Sims, 2011

Minority/poor presence

Increase in minority, decrease in poor

No mention found

Robust

Zapatka & Galvao, 2022

Rent burden, composition

Decrease in burden, increase in Hispanic/foreign-born

No mention found

No mention found

Ahern & Giacoletti, 2022

Wealth transfer

Regressive

2–8% by income

Significant

toof

Pageof

Across these 6 studies:

• Outcome measures:
  • Rent subsidy or discount: 2 studies.
  • Regressive or high-income benefit effects: 2 studies.
  • Minority or racial/ethnic composition or disparity: 2 studies.
  • Rent burden: 1 study.
  • Wealth transfer: 1 study.
  • Some studies covered multiple topics.
• Effect direction:
  • Regressive effects or greater benefits for high-income groups: 2 studies.
  • Positive effect of subsidy on tenure: 1 study.
  • Reduced racial disparity: 1 study.
  • Increased minority presence and decreased poor presence: 1 study.
  • Reduced rent burden with increased Hispanic/foreign-born presence: 1 study.
• Statistical significance:
  • Statistical significance or robustness was reported in 3 studies, based on available full texts or abstracts.
  • We did not find mention of statistical significance in the available full texts or abstracts for the other 3 studies.

Residential Mobility and Tenure Duration

Study

Outcome Measure

Effect Direction

Effect Size

Statistical Significance

Diamond et al., 2019a

Mobility

Decrease in mobility

20% reduction

No mention found

Munch & Svarer

Tenancy duration

Increase in duration

+6 years

No mention found

Mense et al., 2019

Mobility

Decrease (high-income), increase (low-income, short distance)

No mention found

No mention found

Karpestam, 2022

Mobility

Decrease (older units), increase (newer units)

5.7% decrease with 10% increase in income

No mention found

Nagy, 1997

Tenure duration

Increase in duration (stabilized)

No mention found

No mention found

Chapple et al., 2022

Outmigration

Decrease (low socioeconomic status, stabilization), increase (market-rate)

1–2% increase in outmigration

No mention found

toof

Pageof

Across these 6 studies:

• Outcome measures:
  • Mobility: 3 studies.
  • Tenancy or tenure duration: 2 studies.
  • Outmigration: 1 study.
• Effect direction:
  • Decrease only (in mobility): 1 study.
  • Increase only (in tenancy/tenure duration): 2 studies.
  • Mixed effects, with both increases and decreases depending on subgroup or context (for example, by income, unit age, or stabilization status): 3 studies.
• Effect sizes:
  • 20% reduction in mobility: 1 study.
  • 5.7% reduction in mobility with a 10% increase in income: 1 study.
  • Increase of 6 years in tenancy duration: 1 study.
  • 1–2% increase in outmigration: 1 study.
• Statistical significance:
  • We did not find mention of statistical significance in the available full texts or abstracts for any of the 6 studies.

Neighborhood Segregation and Community Composition

Study

Outcome Measure

Effect Direction

Effect Size

Statistical Significance

Sims, 2011

Segregation

Increase in segregation

No mention found

Robust

Chapple et al., 2022

Exclusionary impact

Increase in exclusion

No mention found

No mention found

Chen et al., 2023

Racial disparities

Decrease in disparities

No mention found

No mention found

toof

Pageof

Across these three studies:

• Outcome measures:
  • Segregation: 1 study.
  • Exclusionary impact: 1 study.
  • Racial disparities: 1 study.
• Effect direction:
  • Increase in the measured outcome (segregation or exclusion): 2 studies.
  • Decrease in disparities: 1 study.
• Statistical significance:
  • Statistical significance was described as “robust” in 1 study.
  • We did not find mention of statistical significance in the available full texts or abstracts for the other 2 studies.

References

Jordi Jofre-Monseny, Rodrigo Martínez-Mazza, M. Segu\ (2023).Effectiveness and supply effects of high-coverage rent control policies. Regional Science and Urban Economics

Rebecca Diamond, Timothy J McQuade, Franklin Qian\ (2019).The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco. The American Economic Review

J. Munch, Michael Svarer\ (2001).Rent Control and Tenancy Duration

Andreas Mense, C. Michelsen, Konstantin A. Cholodilin\ (2017).Empirics on the causal effects of rent control in Germany

P. Breidenbach, Lea Eilers, Jan Fries\ (2021).Temporal dynamics of rent regulations – The case of the German rent control. Regional Science and Urban Economics

Andreas Mense, C. Michelsen, K. Kholodilin\ (2019).Rent Control, Market Segmentation, and Misallocation: Causal Evidence from a Large-Scale Policy Intervention. Social Science Research Network

Joseph Gyourko, Peter D. Linneman\ (1989).Equity and efficiency aspects of rent control: An empirical study of New York City

David H. Autor, C. Palmer, Parag A. Pathak\ (2012).Housing Market Spillovers: Evidence from the End of Rent Control in Cambridge, Massachusetts. Journal of Political Economy

Dirk W. Early, John T. Phelps\ (1999).Rent Regulations' Pricing Effect in the Uncontrolled Sector: An Empirical Investigation

G. Fallis, L. B. Smith\ (1985).Price effects of rent control on controlled and uncontrolled rental housing in Toronto: a hedonic index approach

Eilidh Geddes, Nicole Holz\ (2025).Rational Eviction: How Landlords Use Evictions in Response to Rent Control. Social Science Research Network

K. Chapple, Jackelyn Hwang, J. Jeon, Iris Zhang, Julia Greenberg, and 1 more\ (2022).Housing Market Interventions and Residential Mobility in the San Francisco Bay Area. Federal Reserve Bank of San Francisco, Community Development Working Paper Series

Anja M. Hahn, K. Kholodilin, S. Waltl, Marco Fongoni\ (2023).Forward to the Past: Short-Term Effects of the Rent Freeze in Berlin. Management Sciences

J. Weber\ (2017).The Regulation of Private Tenancies - A Multi-Country Analysis

J. Gilderbloom, Linxin Ye\ (2007).Thirty Years of Rent Control: A Survey of New Jersey Cities

H. Donner, Fredrik Kopsch\ (2023).An income-distributional analysis of the rent control subsidy. Journal of Housing and the Built Environment

E. Glaeser, Erzo F. P. Luttmer\ (1997).The Misallocation of Housing Under Rent Control

Teemu Lyytikäinen\ (2006).Rent Control and Tenants' Welfare: The Effects of Deregulating Rental Markets in Finland

Brian J. Asquith\ (2019).Housing Supply Dynamics under Rent Control: What Can Evictions Tell Us?. AEA Papers and Proceedings

Parag A. Pathak, David H. Autory, Christopher J. Palmerz\ (2010).APPLIED ECONOMICS WORKSHOP Business 33610 Spring Quarter 2010

Mats Wilhelmsson, R. Andersson, K. Klingborg\ (2011).Rent Control and vacancies in Sweden

Loïc Bonneval, F. Goffette-Nagot, Zhejin Zhao\ (2015).The impact of rent control: Investigations on historical data in the city of Lyon. Growth and Change

Rebecca Diamond, Timothy J McQuade, Franklin Qian\ (2018).The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco

A. Oust\ (2018).The removal of rent control and its impact on search and mismatching costs: evidence from Oslo

Peter Karpestam\ (2022).Area income, construction year and mobility of renters in Sweden: two hypotheses about the impact of rent control. International Journal of Housing Markets and Analysis

E. Goetz\ (1995).A Little Pregnant

Philipp Großkurth\ (2006).ECONOMIC PAPERS

David P Sims\ (2011).Rent Control Rationing and Community Composition: Evidence from Massachusetts

Rebecca Diamond, Timothy J McQuade, Franklin Qian\ (2019).Who Pays for Rent Control? Heterogeneous Landlord Response to San Francisco’s Rent Control Expansion. AEA Papers and Proceedings

Ruoyu Chen, Hanchen Jiang, Luis E. Quintero\ (2023).Measuring the value of rent stabilization and understanding its implications for racial inequality: Evidence from New York city. Regional Science and Urban Economics

Kasey Zapatka, Juliana de Castro Galvao\ (2022).Affordable Regulation: New York City Rent Stabilization as Housing Affordability Policy. City & Community

K. Kholodilin, Andreas Mense, C. Michelsen\ (2016).Market Break or Simply Fake? Empirics on the Causal Effects of Rent Controls in Germany

K. Kholodilin, Sebastian Kohl\ (2020).Does Social Policy through Rent Controls Inhibit New Construction? Some Answers from Long-Run Historical Evidence

D. Flaming, P. Burns, Michael Matsunaga, Mirna Ponce, K. Baar, and 3 more\ (2009).Economic Study of the Rent Stabilization Ordinance and the Los Angeles Housing Market

J. Nagy\ (1997).Do Vacancy Decontrol Provisions undo Rent Control

K. Kholodilin, Fernando A. López, David Rey Blanco, Pelayo Gonzalez Arbués\ (2022).Lessons from an Aborted Second-Generation Rent Control in Catalonia. Social Science Research Network

K. Ahern, M. Giacoletti\ (2022).Robbing Peter to Pay Paul? The Redistribution of Wealth Caused by Rent Control. Social Science Research Network

William L. C. Wheaton\ (1981).Housing policies and urban “markets” in developing countries: The Egyptian experience☆

Max Gardner, Brian Asquith\ (2024).The Effect of Rent Control Status on Eviction Filing Rates: Causal Evidence From San Francisco. Housing Policy Debate

Yeon Hur, Hanming Gu, Kabsung Kim\ (2025).Do the Right to Lease Renewal and the Rent Cap Curb Apartment Rent Increases?: Dissecting Rent Dynamics in a Regulated Housing Market. Korean Association for Housing Policy Studies

Download BIBDownload RISDownload TXT

Report

Status

Gather sources

989 sources found

Details

Screen sources

40 sources included

Details

Extract data

200 data points extracted

Details

Generate report

Save PDF

BIBLaTeX, ZoteroRISZotero, MendeleyTXTAPA BibliographyPDFPDFDOCXMicrosoft Word

Chat

Got some follow-up questions?

Sign up or sign in to chat with this report.

Back

Housing Market Spillovers: Evidence from the End of Rent Control in Cambridge, Massachusetts

David H. Autor, C. Palmer, Parag A. Pathak

Journal of Political Economy·

2012·

210 citations

SourceDOI

Plain textPDFFigures

Searching for PDF

Unable to find PDF from source

Back

Study Design Type

Quasi-experimental; Difference-in-differences; Econometric study

Geographic Setting and Context

- Country: United States - City/metropolitan area: Cambridge, Massachusetts - Specific neighborhoods or regions: Not specified - Contextual details: The study focuses on the effects of rent control elimination in Cambridge, with a focus on price appreciation and investment booms.

Rent Control Policy Details

- Year of policy implementation: 1970 - Specific type of rent control: First-generation - Key policy mechanisms: Strict caps on rent increases, across-the-board rent increases for operational costs, no vacancy decontrol - Scope of application: Non-owner-occupied rental housing built before 1969 - Exceptions/Special conditions: Excludes structures built after January 1, 1969, owner-occupied condominiums, nonresidential structures converted to rental properties after this time; Removal Permit Ordinance restricted unit removal and conversions

Policy Impact Outcomes

- Rental prices: Median rents rose by 40% for formerly controlled units and by 13% for never-controlled units between 1994 and 1997. - Housing supply: Stock of houses decreased by 6%, stock of condominiums increased by 32%, with 45% due to conversions. - Land values: $2.0 billion of $7.7 billion in property appreciation attributed to rent control removal. - Maintenance investments: Building permits increased by 20%, annual expenditures roughly doubled. - Tenant stability: Turnover rates increased by 5.4 percentage points at formerly controlled units. - Demolition rates: Not mentioned. - Housing market dynamics: Appreciation was largely due to spillover effects on never-controlled properties.

Key Quantitative Findings

- Rent-controlled properties were valued at a 45-50% discount relative to never-controlled properties before decontrol. - Assessed values of rent-controlled properties rose by 18-25% relative to never-controlled properties after decontrol. - The indirect effect of rent control removal on never-controlled properties contributed $1.7 billion to the total $2.0 billion increase in Cambridge property values. - The direct effect on decontrolled units accounted for $300 million of the total increase. - More than half (55%) of the capitalized cost of rent control was borne by owners of never-controlled properties. - The indirect effect on never-controlled properties accounted for more than half (56%) of the decontrol-induced increase in housing stock value.

We measure the capitalization of housing market externalities into residential housing values by studying the unanticipated elimination of stringent rent controls in Cambridge, Massachusetts, in 1995. Pooling data on the universe of assessed values and transacted prices of Cambridge residential properties between 1988 and 2005, we find that rent decontrol generated substantial, robust price appreciation at decontrolled units and nearby never-controlled units, accounting for a quarter of the $7.8 billion in Cambridge residential property appreciation during this period. The majority of this contribution stems from induced appreciation of never-controlled properties. Residential investment explains only a small fraction of the total.

I. Introduction

Spillovers from the attributes and actions of neighborhood residents onto the value of surrounding properties and neighborhoods are central to the theory of urban economics and the development of efficient housing policy ðFujita 1991; Glaeser and Gyourko 2009Þ. Credibly identifying and quantifying these external effects, however, pose a significant empirical challenge because key features of the housing market equilibriumin particular, who lives where, the quality and quantity of housing, the levels of local public goods and amenities, and what prices prevail-are all determined simultaneously in equilibrium. 1 This paper exploits an unusual, large-scale policy change, the elimination of rent control in Cambridge, Massachusetts, in 1995, to quantify the capitalization of residential housing market externalities onto the value of residential real estate. From December 1970 through 1994, all rental units in Cambridge built prior to 1969 were regulated by a far-reaching rent control ordinance that placed strict caps on rent increases and tightly restricted the removal of units from the rental stock. The legislative intent of the rent control ordinance was to provide affordable rental housing, and at the eve of rent control's elimination in 1994, controlled units typically rented at 40-plus percent below the price of nearby noncontrolled properties, though maintenance and amenities in controlled units tended to be subpar ðSims 2007Þ. 2 The policy change that provides the identifying variation for our study is the swift elimination of Cambridge's rent control law via a statewide ballot initiative. In November 1994, the Massachusetts electorate passed a referendum to eliminate rent control by a narrow 51-49 percent margin, with nearly 60 percent of Cambridge residents voting to retain the rent control ordinance. Thus, rent decontrol in Cambridge, which commenced only 2 months after the November 1994 referendum, was voted into law by Massachusetts cities and towns that had never experienced rent control, while, ironically, the three Massachusetts municipalities with active rent control regimes-Cambridge, Boston, and Brookline, each of journal of political economy which voted to maintain rent control-were overruled by the statewide majority. 3 longside its swift and largely unanticipated elimination, two unusual features of Cambridge's rent control ordinance make it well suited to credibly identify the effects of rent control on residential housing markets. First, because the rent control ordinance applied to only a fixed, nonexpanding set of residential units-specifically, non-owner-occupied rental houses, condominiums, or apartments built prior to 1969-controlled and never-controlled units stood side by side in Cambridge neighborhoods on the eve of rent control removal, thus offering a tight temporal and geographic framework for assessing the impact of the law on residential property prices. 4 Second, although roughly a third of residential units were controlled prior to elimination ðsee fig. 1Þ , this fraction frequently exceeded 60 percent in neighborhoods that had older housing stocks and a substantial share of renters at the time of rent control's enactment in 1970. This sizable cross-neighborhood variation allows us to assess localized price effects by comparing pre-and postremoval price appreciation among both decontrolled and never-controlled properties in neighborhoods that differed in their "rent control intensity," that is, the share of residential units that were controlled.

Our conceptual model and empirical work distinguish two channels through which rent decontrol may affect the market values of residential properties. The first, which we term the direct effect, reflects the capitalization of landlords' newfound ability to charge market rents. In the absence of any change in residential investments or neighborhood characteristics-and assuming that price controls were binding-rent control removal should directly raise the ownership value of formerly controlled properties by uncapping rents and, simultaneously, increasing the returns to landlord investments. The second channel, which we term the indirect effect, encompasses the multiple complementary mechanisms by which rent decontrol may affect the desirability of surrounding properties: owners renovate and modernize decontrolled units, raising their rental values; affluent tenants who particularly value these amenities rent these units as incumbents depart in the face of rising prices; higherincome tenants move into nearby never-controlled properties, attracted by the amenities of an improved housing stock and more affluent neighbors; and property owners make further investments in both decontrolled and never-controlled units as overall tenant income levels rise. 5 istinct from the "direct" effect of decontrol, which by definition operates only on formerly controlled properties, the indirect channel may affect the market value of both decontrolled and never-controlled properties by increasing the desirability of the neighborhoods in which they are located. While our analysis does not allow us to further decompose the indirect effect into its constituent components ðinvestment, reallocation, and the complementarities between the twoÞ, historical evidence suggests that each of these channels was relevant. Because Cambridge's Rent Control Board was unlikely to grant rent increases following property improvements, it was widely perceived that rent control muted owners' incentives to maintain and improve controlled properties. 6 Consistent with this view, Sims ð2007Þ finds that chronic maintenance problemssuch as holes in walls or floors, chipped or peeling paint, and loose railings-were more prevalent in controlled than in noncontrolled units during the rent control era and that this differential fell substantially with rent control's elimination. The end of rent control also spurred substantial renovations and repairs of ill-maintained decontrolled units ða direct effectÞ; the complementarities among tenant incomes, neighborhood amenities, and the quality of the housing capital stock should raise the return to investments at both decontrolled and never-controlled units ðan indirect effectÞ. tenant turnover. Cambridge's rent control law was intended to enable less affluent tenants to reside in units that would command high rents under a market allocation, particularly the dense neighborhoods proximate to Cambridge's major universities, commercial centers, and transportation hubs. While there was no formal mechanism to allocate controlled units to low-income households, limited quantitative evidence indicates that less affluent residents and students were overrepresented in controlled units, though a significant number of units were also occupied by wealthy professionals. 7 As we show below, exit rates from formerly controlled units spiked in the years immediately following rent decontrol. And given the substantial accompanying increases in rents, it is likely that the new cohorts of renters were significantly more affluent than the tenants they replaced. Our analysis will capture the net effect of these potentially mutually reinforcing channels on the market value of Cambridge residential real estate.

Regulations are widespread in housing markets, and rent controls are arguably among the most important historically ðFriedman and Stigler 1946; Glaeser and Gyourko 2009Þ. Because they directly manipulate the price mechanism, they are likely to reshape the allocation of residents to locations, the incentives for investment and maintenance of controlled units, and the supply, demand, quality, and allocation of units in the noncontrolled sector. The modern era of US rent controls began as a part of World War II-era price controls and as a reaction to housing shortages following demographic changes immediately after the war ðFetter 2013Þ. While the prevalence of rent control as a housing market policy has decreased since this period, rent control and rent stabilization plans are still in place in many US and European cities ðArnott 1995Þ. New York City's system of rent regulation affects at least 1 million apartments, while cities such as San Francisco, Los Angeles, Washington, DC, and several California and New Jersey cities have various forms of rent regulation. Rent control remains a topic of active debate among affordable housing advocates.

The early empirical literature on rent control focuses on its effects on the supply of housing services ðOlsen 1972Þ and the incentives of landlords to invest in building quality ðFrankena 1975; Gyourko and Linneman 1989Þ. A second strand of this literature examines how belowmarket rents may encourage individuals to spend effort to obtain cheap 7 A 1998 study commissioned by the City of Cambridge found that sitting residents of formerly controlled units had mean annual earnings in 1997 of $35,650 vs. $43,630 among tenants of market rate units and $41,340 among tenants of formerly controlled units who had taken residence after rent control removal ðAtlantic Marketing Research 1998Þ. Sims ð2007Þ calculates that 67 percent of residents of rent-controlled units in Boston, Brookline, and Cambridge were in the bottom two quartiles of the income distribution. At the same time, blacks were substantially underrepresented in controlled units.

housing, leading to a misallocation of housing ðSuen 1989; Glaeser and Luttmer 2003; Sims 2011Þ. Fallis and Smith ð1984Þ examine how the impact of rent control on the uncontrolled sector depends on the allocation mechanism in the controlled sector. Wang ð2011Þ investigates the impact of privatization of housing that was owned and allocated by the state in urban China. Her analysis, like ours, shows that the degree of misallocation of assets prior to privatization affects the expected change in prices.

Sims ð2007Þ undertakes the first empirical analysis of the end of rent control in Massachusetts, exploring its impacts on the supply of rental properties and their rental prices. Sims shows that the elimination of rent control spurred substantial rent increases in Massachusetts towns that had binding rent control laws in 1994 ðBoston, Brookline, and Cam-bridgeÞ and led to significant increases in the quality and quantity of rental housing available. In contrast to Sims's work, we analyze rent control's effect on the market value ðrather than rental pricesÞ of the entire residential housing stock ðnot simply rental unitsÞ in Cambridge and distinguish its effects on decontrolled properties and never-controlled properties. 8 ur work is also related to studies of neighborhood revitalization and gentrification, both of which may generate spillover benefits to surrounding areas ðIoannides 2003; Schwartz et al. 2006; Rossi-Hansberg, Sarte, and Owens 2010; Guerrieri, Hartley, and Hurst 2013Þ. Studies by Linden and Rockoff ð2008Þ and Pope ð2008Þ of the housing market impacts of the arrival of registered sex offenders into a neighborhood consider allocative externalities in residential housing. Recent interest in measuring external effects in housing has been spurred in part by historically high levels of foreclosures and the concern for their impact on immediate neighbors and neighborhoods ðHartley 2010; Campbell, Giglio, and Pathak 2011; Mian, Sufi, and Trebbi 2011Þ. 9 Our analysis draws on a uniquely detailed geographic and economic database sourced from Cambridge administrative records that enumerates the exact location of all rent-controlled units, the assessed value of each house and condominium in 1994 and 2004, the transacted price of each residential property sold between 1988 and 2005, the movement of properties across various residential and nonresidential uses ðe.g., houses that were converted to condominiumsÞ, and the permitted investment expenditures at each residential location. We additionally use 10 years of Cambridge city census data to document the rapid turnover of residents of formerly controlled units following the end of rent con-trol. These sources permit direct estimation of changes in residential real estate prices induced by rent decontrol.

We find compelling evidence that the elimination of rent control raised the market values of both decontrolled and never-controlled properties. Our main estimates imply that during the rent control era, rentcontrolled properties were valued at a discount of about 45-50 percent relative to never-controlled properties with comparable characteristics in the same neighborhoods and that their assessed values rose by 18-25 percent relative to never-controlled properties following rent decontrol. This differential appreciation should primarily reflect the direct effect of rent decontrol on the market value of formerly controlled units generated by the potential for owners to charge market rents, the option to convert rental units into condominiums, and the flow of returns from associated capital investments.

To assess whether rent control density affected the desirability of neighborhoods over and above its direct effect on controlled properties, we next calculate a rent control exposure measure for each residential unit that is equal to the fraction of other residential units within a 0.20-mile radius that were subject to rent control as of 1994. A central finding is that post-decontrol price appreciation was significantly greater at units that had a larger fraction of formerly controlled neighbors: residential properties at the 75th percentile of rent control exposure gained approximately 13 percent more in assessed value following decontrol than did properties at the 25th percentile of exposure. This differential appreciation of properties in rent control-intensive locations was equally pronounced among decontrolled and never-controlled units, suggesting that rent control removal spurred overall gains in neighborhood desirability.

These findings are robust to many alternative measures of rent control intensity, to rich controls for property-level characteristics ðsuch as age, lot size, and number of bedrooms and bathroomsÞ, and to the inclusion of detailed geographic fixed effects and neighborhood trends that allow price levels to vary across Cambridge neighborhoods and to trend over time within them. Data on transaction prices for all properties sold in Cambridge between 1988 and 2005, which provide an alternative source for measuring changes in market values, yield estimates of spillover effects comparable to those found using the assessor's data.

One channel through which the removal of rent controls may have raised Cambridge housing values is spurring additional capital investments. Using administrative data on residential expenditures permitted by the Cambridge Inspectional Services, we find that aggregate annual permitted building expenditures increased dramatically for both houses and condominiums after 1994, rising from $21 million per year between 1991 and 1994 to $45 million per year between 1995 and 2004. More-over, the incidence of permitting-though not investment expenditures per unit-rose differentially at formerly controlled properties in the years immediately following rent control removal. But the total value of Cambridge residential investments in these 10 years was less than one-quarter as large as the estimated increment to Cambridge residential housing values induced by rent control removal, suggesting that the allocative rather than the investment channel is the more important explanation for the post-1994 rise in the market value of never-controlled properties.

The economic magnitude of the effect of rent control removal on the value of Cambridge's housing stock is large, contributing $2.0 billion of $7.7 billion in Cambridge property appreciation in the decade between 1994 and 2004. Of this total effect, only $300 million is accounted for by the direct effect of decontrol on formerly controlled units ðholding exposure constantÞ, while $1.7 billion is due to the indirect effect. Notably, the majority of this indirect effect ð$1.1 of $1.7 billionÞ stems from the differential appreciation of never-controlled units. When both direct and indirect effects are combined, our estimates imply that more than half ð55 percentÞ of the capitalized cost of rent control was borne by owners of never-controlled properties.

The paper proceeds as follows. Section II provides additional details on the enactment, enforcement, and removal of rent control in Cambridge. Section III describes a simple model of housing markets in the presence of rent control to guide our empirical analysis ðApp. A contains the modelÞ. Section IV describes data sources and our empirical strategy. Section V presents our main results using property assessments, while Section VI presents results on the time path of the capitalization of rent decontrol using transaction prices. Section VII reports on our investigation of permitting and investment activity, and Section VIII considers economic magnitudes. We conclude with a discussion of areas for further investigation.

A. Rent Control Adoption and Elimination

In 1970, the Massachusetts state legislature enacted a statute allowing cities and towns with populations over 50,000 to implement rent control to "alleviate the severe shortage of rental housing." 10 Boston, Brookline, Cambridge, Lynn, and Somerville each adopted a rent control plan, with Cambridge moving first in 1970 and keeping the ordinance longer than any other city. Lynn repealed its plan in 1974 and Somerville in 1979. Boston allowed for decontrol of vacant units in 1976, and Brookline began to phase out its system prior to the statewide repeal, though both cities still had a significant number of controlled units in 1994 ðCantor 1995Þ. 11 In Cambridge, rent control was seen as an integral part of the city's affordable housing program.

Cambridge's initial rent control policy adopted in 1970 applied to all non-owner-occupied rental housing built before 1969. It did not apply to structures built after January 1, 1969, to owner-occupied condominiums, or to nonresidential structures converted to rental properties after this time. Oversight of the rent control law rested with the Cambridge Rent Control Board, whose official charter was to ensure that landlords obtained a fair net operating income. The board established maximum allowable rents for each controlled property with the aim of fixing landlord net operating income at inflation-adjusted 1967 levels. In the 1970s and 1980s, the board authorized a series of across-the-board rent increases ranging from 1.15 to 3.1 percent, intended to cover increases in heating costs, operating costs, and property taxes. 12 Landlords could also apply to raise prices above the scheduled increases, but these variances were rarely sought or granted in practice, in part because the application required supporting petitions, extensive legal documentation, and significant time investment. 13 istinct from many cities, Cambridge's rent control policy did not allow for so-called vacancy decontrol, whereby controlled rental units were returned to market rate rents after protected tenants moved out. Landlords therefore faced an incentive to remove units from the rental stock, which they accomplished by converting substantial numbers of rental units to condominiums and selling them to owner-occupants. To prevent the controlled rental stock from being depleted, in 1979 the city council passed the Removal Permit Ordinance, which substantially restricted the removal of controlled units from the rental stock and complicated the conversion of controlled units into owner-occupied condominiums. 14 he development that ultimately led to rent control's elimination was the Cambridge Small Property Owners Association's successful effort to 11 See Epple ð1988Þ for a game-theoretic model of communities' decisions to adopt rent control.

12 All documents from the Rent Control Board are available in the archives of the Cambridge Historical Commission. 13 A legendary incident involves Harvard philosophy professor Robert Nozick extracting a settlement of over $30,000 in the 1980s from his landlord, famed classicist and novelist Eric Segal, for overcharging rent, described in Tucker ð1986Þ.

14 This ordinance required proof that removal would not aggravate the housing shortage and would "benefit the persons sought to be protected" by the rent control statute ðCantor 1995Þ. The ordinance was subsequently amended following difficulties with enforcement, which were made salient by the fate of the so-called condo martyrs: owners who were prosecuted for occupying their own controlled properties before the completion of a conversion.

place rent control on the statewide ballot in 1994. Putting rent control to a statewide vote diluted the strong support that rent control enjoyed in the three municipalities with extant rent control ordinances ðBoston, Brookline, and CambridgeÞ. Rent control was eliminated by a slim 51-49 percent margin in November 1994, despite nearly 60 percent of Boston, Brookline, and Cambridge voters voting to retain the current regime. Just 2 months later in January 1995, a majority of properties were decontrolled. A last-minute legislative compromise, however, allowed disabled, elderly, and low-income renters to retain their current units at their controlled rents for up to 2 years. Though only a small share of residents received rent control extensions, this compromise likely created some uncertainty about whether decontrol was final, at least until the grandfathering period expired in 1997 with no further controls in place. 15

B. The Post-decontrol Regime

The elimination of rent control catalyzed a series of rapid changes in the Cambridge rental market, beginning with rising rents. A 1998 survey commissioned by the City of Cambridge ðAtlantic Marketing Research 1998Þ found that nominal Cambridge median rents rose by 40 percent between 1994 and 1997 for tenants of formerly controlled units who either remained at these units or moved to other noncontrolled units. Median rents rose by only 13 percent for sitting tenants of never-controlled units in the same time period.

Rising rents spurred a sharp increase in resident turnover at formerly controlled units after 1994, which we document by constructing a panel of all Cambridge adults aged 17 and older by street address using city voter registration records for the years 1991-2000. 16 On average, 26.9 percent of Cambridge residents changed locations annually, with the highest turnover rates found among apartment residents ð33.5 percentÞ, followed by residents of condominiums ð29.7 percentÞ and houses ð23.2 percentÞ.

We assess whether turnover rates at formerly controlled units rose differentially after 1994 by fitting linear probability models of the following form:

where NEW ijt is an indicator equal to one if resident i in unit j in year t was not present in that unit in the prior year. In this model, RC j is an indicator equal to one if unit j was rent controlled in 1994, g g is a vector of 1990 census block group dummies, d t is a vector of year dummies, and Post t is an indicator for years 1995 onward. Prior to 1995, residents of controlled units were not significantly more likely to turn over than residents of noncontrolled units. 17 Following decontrol, the turnover differential between formerly controlled and never-controlled units rose by 5.4 percentage points, with an even larger increase at condominiums ðtable 1Þ.

Figure 2 depicts the evolution of this turnover differential using a variant of equation ð1Þ in which the rent control indicator is interacted with a set of year dummies. Turnover rates at decontrolled units spiked by 4 percentage points relative to never-controlled units in the first year of decontrol and continued to climb to 10 percentage points over the next 3 years. Thus, the process of resident reallocation and neighborhood change spurred by decontrol took multiple years to unfold. Interestingly, figure 2 also shows that turnover rates at never-controlled units changed little following decontrol.

A sharp increase in residential property investments also followed the end of rent control. The number of building permits issued per residential unit for improvements and new construction increased by approximately 20 percent after 1994, and annual permitted expenditures roughly doubled in real terms ðsee App. table A1Þ . Elimination of the Removal Permit Ordinance allowed a substantial number of decontrolled houses, apartments, and nonresidential units to be converted to condominiums. From 1994 to 2004, Cambridge's stock of residential houses decreased by 6 percent, while the stock of condominiums increased by 32 percent, with 45 percent of this increase accounted for by conversion of houses to condominiums ðApp. table A2Þ . 18 At the same time, the fraction of residential units available as rental properties rose by 6 percentage points ðSims 2007Þ. 17 Subsequent columns in table 1 reveal that this result is driven by composition. If we focus only on apartments and condominiums, residents of controlled units were significantly less likely to turn over than residents of noncontrolled units-consistent with the idea that controlled units had scarcity value. Residents of controlled houses, by contrast, were significantly more likely to turn over than residents of noncontrolled houses, but this likely reflects the fact that most noncontrolled houses were owner-occupied whereas controlled houses were renter-occupied. 18 These calculations use the Cambridge assessor's databases from 1995 and 2005, reflecting the status of properties in 1994 and 2004, respectively. We count each unit in multifamily houses separately to meaningfully compare the supply of housing across different structure types and in different periods. The stock of units in houses in Cambridge decreased from 14,722 in 1994 to 13,861 in 2004 and the stock of condominiums rose from 7,220 to 9,561 units. 1992-2000. The figure plots coefficients on RC Â Year variables from an event-study regression in which the dependent variable is an indicator equal to one if the resident was not present in the current Cambridge unit in the prior year ðand zero otherwiseÞ. RC is an indicator for a location that was rent controlled in 1994. This specification includes an RC main effect, year controls, structure type dummies, and geographic fixed effects for the 91 block groups in the 1990 census containing addresses listed in the Cambridge city census. The 95 percent confidence intervals are constructed from robust standard errors clustered by block group. The vertical line in 1994 indicates the year preceding rent control removal. This combination of sizable rent increases, rapid turnover of incumbent renters, rising residential investment, and outward shifts in the supply of both condominiums and rental properties was likely in net to have changed the quality of the Cambridge residential housing stock, the allocation of residents to neighborhoods, and the availability of residential units for both rent and sale.

III. The Direct and Indirect Effects of Rent Control

Appendix A presents a stylized model of the housing market, summarized here, that considers the relationship between rent control and prices of both controlled and noncontrolled properties. In the model, a city consists of N neighborhoods with a continuum of locations in each neighborhood. Potential residents choose locations to maximize utility defined over consumption of housing services, a nonhousing composite good, and local amenities. Residents have identical preferences and differ only in their income levels. Profit-maximizing landlords choose the level of maintenance at each location, and this level is increasing in the price of housing services.

We assume that amenities in a neighborhood depend on the housing maintenance levels and the income distribution of residents in the neighborhood, where higher maintenance and higher-income neighbors are also more desirable and hence contribute more to neighborhood amenities. This formulation creates positive feedback from the extent of maintenance, residents' income, and neighborhood amenities. In the free-market equilibrium ðwith no rent controlsÞ, rents are higher in neighborhoods with greater amenities as a result of higher maintenance and the presence of higher-income neighbors.

We consider the imposition of rent controls at the initial free-market equilibrium by assuming that a rent control authority caps the rent of some units in a neighborhood at below their free-market level. Since landlords choose maintenance levels facing a regulated price, maintenance levels and hence housing services are lower at controlled units. The combination of reduced rents and lower maintenance has one of two effects on incumbent residents: either they are sufficiently compensated by reduced rents so that they remain at their current locations, although the bundle of maintenance and amenities is not optimized for their income levels, or, alternatively, they choose to relocate to areas with higher amenities and higher rents. In the latter case, they will be replaced by residents who prefer lower housing services, that is, those with lower incomes. 19 The average income at controlled locations therefore weakly declines following the imposition of rent control.

Since neighborhood amenities are a function of the maintenance of all units in a neighborhood and the neighborhood income distribution, the levels of amenities at noncontrolled locations in these neighborhoods-as well as maintenance and rents-are also impaired by rent control. This in turn causes lower-income residents to move into noncontrolled locations. Thus, rent control causes inefficiently low maintenance and misallocation of residents at both controlled and noncontrolled locations within a neighborhood.

Decontrol unwinds these effects. Prices rise directly because of the lifting of the cap and indirectly because of improved maintenance and increased production of local amenities throughout the neighborhood. At noncontrolled locations, the price increase will be greater in neighborhoods where a larger fraction of locations were controlled, where the capped price ceiling was set further below the market price level, and where controls induced larger resident misallocation relative to the free-market setting. The lifting of controls allows an additional, direct price increase at formerly controlled locations.

The model also offers a simple welfare interpretation of any direct and indirect price effects of rent decontrol. Price increases at decontrolled locations reflect three forces: a mechanical "uncapping" effect, which reflects a transfer from renters to owners; a price increase reflecting improved maintenance, which generates increased landlord surplus net of the resource cost of maintenance; and a price increase reflecting greater neighborhood amenities due to improvements in maintenance and changes in resident types nearby. While the latter two effects reflect economic gains, the first does not. The price increase at decontrolled locations is therefore likely to substantially exceed the economic gains from decontrol at these locations.

Induced price increases at noncontrolled locations following decontrol reflect the capitalization into house values of two of these three forces: improved maintenance ðor, more generally, housing investmentsÞ and greater neighborhood amenities ðboth due to sorting and capital improvements at other propertiesÞ. Therefore, the increase in prices at noncontrolled locations, net of the additional resource costs expended on maintenance and improvements, can be used to assess the external effects of decontrol, that is, the spillovers. We quantify these spillovers below by estimating the increase in market value of never-controlled units and netting out the components plausibly attributable to investment.

IV. Data and Measurement

We briefly discuss our data sources and measurement of rent control intensity in this section, with further details in online Appendix B. journal of political economy

A. Cambridge Real Estate

There are approximately 15,000 taxable parcels of land in the city of Cambridge organized into unique geographic units known as "map-lots." The foundation for our data set is a snapshot of the entire universe of residential real estate from the 1995 Cambridge Assessor's File, from which we construct the residential housing structures file. 20 Each record includes the map-lot identifier, address, owner's name and address, usage, and property tax assessment as of January 1994. Usage categories are designated as commercial or residential, and residential categories are further subdivided into condominiums; single-family, two-family, and three-family houses; multi-unit apartment complexes; and mixed residential-commercial structures. In calculating rent control intensity below, we treat any usage code in which individuals are likely to live as a residential structure. Our analysis of assessed values and transactions is limited to houses and condominiums, which make up the market for residential real estate.

We identify rent-controlled properties from historical records of the Cambridge Rent Control Board obtained via a Freedom of Information Act ðFOIAÞ request. 21 We merge rent control structures to the assessor's file using the map-lot identifier and address information coded in the Rent Control Board file. Rent-controlled records that could not be matched via map-lot identifiers were hand-matched to the corresponding street address. Owing to limitations of the Rent Control Board data, it was often not possible to determine which specific units in a multi-unit building were controlled. This creates a potential econometric pitfall: if we were to inadvertently code some controlled units as never controlled, our data analysis could erroneously detect spillovers that reflect nothing more than appreciation of formerly controlled units after decontrol. To be conservative, we code all units on a map-lot as rent controlled if any unit at that map-lot was controlled in 1994. It is therefore very unlikely that there are controlled units that we fail to capture. Conversely, when measuring the rent control intensity of a given geographic area, we calculate the fraction of residential units-rather than structures-that are rent controlled. 22 his is also conservative in that it prevents us from overestimating units' exposure to other controlled properties.

Figure 1 illustrates the prevalence of rent control in Cambridge, with dark circles indicating controlled properties. In 1994, 22 percent of all residential structures and 38 percent of residential units were subject to rent control. The dense neighborhoods close to the two major universities and proximate to the subway that bisects Cambridge from east to northwest contain high concentrations of renters and multi-unit structures and thus had relatively high rent control intensity. The largely owneroccupied area of southwestern Cambridge features a higher fraction of single-unit houses and hence had relatively low rent control intensity. It bears emphasis that our statistical analysis abstracts from these gross geographic differences in rent control intensity by comparing changes in residential prices among properties that differ in their proximity to controlled units but lie within relatively small neighborhoods.

We append two databases to analyze the impact of rent decontrol on market capitalization, the first enumerating property assessments and the second enumerating real estate transactions. The 1995 and 2005 Cambridge Assessor's Files, which report property valuations from 1994 and 2004, provide the assessed appreciation of each extant property from the year prior to rent decontrol to 9 years thereafter. The second is a commercial database provided by the Warren Group, which enumerates all changes in ownership of residential properties for the years 1988-2005. Sourced from records of deeds, these data log each real estate transaction, including sale price, address, map-lot, number of bedrooms and bathrooms, lot size, year built, and property type. We exclude commercial properties such as apartment buildings from the analysis because such sales are rare and transact at heterogeneous prices that are in some cases extremely high.

Assessments and transactions provide complementary means to measure the capitalization of rent control's end. Assessments, our preferred measure, contain the universe of residential properties along with assessed market values at two points in time, immediately prior to rent control removal and 10 years later. Assessments may offer a lagging indication of residents' changing willingness to pay for locations, however, and could differ from market valuations because of discretionary aspects of the assessment process. The sales data, in contrast, include both market prices and a rich set of property characteristics for locations where transactions take place and, because they are available annually, provide a clearer picture of the trajectory of property price changes. Only a small percentage of residential units transact each year, however, and hence the sales data contain information on an incomplete and potentially nonrepresentative set of residential units ðwe subsequently analyze whether rent control affected the composition of transacted propertiesÞ.

Table 2 presents descriptive statistics for the assessed Cambridge residential houses and condominiums used in our analysis, comprising 15,475 properties in 1994 and 17,505 in 2004. 23 Slightly more than half of these properties are houses. Rent-controlled properties account for 29 percent of all residential properties, with condominiums making up the substantial majority. Because the vast majority of Cambridge houses were and are owner-occupied, only 12 percent of houses were ever subject to rent controls. 24 House prices rise substantially in real terms during our sample period: the average 1994 assessed value of a decontrolled condominium is $116,000, while it is $351,000 in 2004-an increase of 111 log points. 25 ouses typically have higher assessed values than condominiums, and in both periods, decontrolled houses and condominiums have lower values, on average, than never-controlled houses.

B. Measuring Rent Control Intensity (RCI)

Gauging each residential property's rent control exposure requires a metric that specifies which nearby units should be counted in the unit's reference set-that is, to which units it is "exposed"-and how the rent control status of these reference units should be aggregated into an exposure index. For most analyses, we calculate the rent control status of the surrounding units to which a given property i is exposed by summing the number of controlled units within a surrounding geography g and dividing it by the sum of all residential units J g ðcontrolled and noncon-trolledÞ in that geography:26

In a subsequent sensitivity analysis, we calculate each unit's rent control exposure as an exponentially declining function of its distance from all other controlled and never-controlled properties in the city.

The second input into the exposure measure is the choice of a surrounding geography. One potential set of geographies is supplied by the US Census Bureau, which subdivides the area of cities into three increasingly fine geographic units: tracts, block groups, and blocks, of which there are 30, 89, and 587, respectively, in Cambridge containing at least one assessed house or condominium. 27 While these predefined census geographies have the virtue of allocating Cambridge land parcels into exhaustive, mutually exclusive geographic units, they have two substantial drawbacks for our analysis. One is that the census geographies do not necessarily correspond to any specific notion of neighborhoods or proximity. For example, census blocks frequently divide streets down the center, so that units on opposite sides are assigned to different blocks, which is clearly undesirable for measuring spillovers from nearby properties. The second is intrinsic to any allocation of geography into nonoverlapping parcels: units closer to the perimeter of a geography are treated differently from units located in its center. For example, for a residential unit located on the northern edge of a geography, its neighbors 50 feet to its south will contribute to the unit's rent control exposure measure whereas its neighbors 50 feet to its north will not. By contrast, for a unit located in the center of a geography, its equidistant neighbors contribute equally to its rent control exposure measure.

To avoid both drawbacks of using fixed geographies, our preferred measure of a unit's rent control exposure is the fraction of residential units within a fixed straight-line radius of 0.10, 0.20, and 0.30 mile of that unit that were controlled as of 1994. This radius exposure construct nonprejudicially selects the residential units that are physically closest to the reference unit. 28 To provide a feel for the area encompassed by these radii, figure 1 plots concentric rings of appropriate scale overlaid on the Cambridge map.

Our main estimates are based on RCI measured at a radius of 0.20 mile, which corresponds to about 0.13 square mile-an area larger than a block group but smaller than a tract in our sample. For the typical residential property, 34 percent of the surrounding units within a 0.20-mile radius are rent controlled. As shown in table 2, condominiums are in neighborhoods with more rent control than houses, and both decontrolled houses and condominiums tend to be in more rent control-intensive neighborhoods than their never-controlled counterparts. For instance, in 1994, 32 percent of units surrounding a typical never-controlled condominium are controlled, compared to 45 percent for decontrolled condominiums. There is also considerable cross-sectional variation in rent control intensity. Across all assessed properties, the standard deviation of RCI measured at 0.20 mile is 17 percentage points, and the range of the RCI measure spans from 0 to 72 percent.

V. Capitalized Effects of Rent Decontrol: Evidence from Assessments

Our illustrative model suggests that the capitalization of rent decontrol should accrue through three channels: the direct effect on decontrolled properties of the elimination of price controls and condominium conversion restrictions and associated investments; the indirect effect of decontrol on the desirability of neighborhoods in which controlled properties were located, stemming from improvements in neighborhood amenities-for example, better upkeep, more desirable neighbors-and potentially affecting the market value of both decontrolled and never-controlled properties; and finally, broader increases in the desirability of Cambridge as a residential location, which may accrue citywide.

Our econometric model recognizes each of these channels. We fit equations of the form

where Y A igt is the real assessed value of property i in neighborhood g in year t, g g are fixed effects representing different geographies, d t are year effects, and X i are property characteristics such as housing type ðcondominium, single-family, two-family, or three-family houseÞ. The dummy variable RC i is equal to one for properties that were rent controlled in 1994 ðprior to the law's repealÞ, while the Post indicator is equal to one for 2004. Of central importance to the analysis, the variable RCI i measures the fraction of units nearby to i that were controlled as of 1994. Our main specifications code "nearby" units as those within a 0.20-mile radius of a given property, but we subsequently explore alternative definitions. Recognizing that real estate prices of nearby properties are not independent, we generally cluster the standard errors at the level of Cambridge block groups.

The coefficient r 1 estimates the direct effect of rent control removal on the assessed value of formerly controlled properties by contrasting the change in value of controlled versus never-controlled properties following the end of rent control, holding constant unit characteristics, cross-neighborhood differences in residential real estate prices and over time, and citywide changes in residential real estate prices. The coefficient r 2 estimates the indirect effect of rent decontrol on the value of decontrolled and never-controlled properties by contrasting changes in the value of units in geographies with high rent control intensity relative to those with low rent control intensity, again holding constant property characteristics, neighborhood effects, and time effects. Finally, any effects of decontrol that accrue citywide-that is, are not limited to decontrolled properties or the neighborhoods in which they were located-are absorbed by the time effects d t . Since these time effects soak up any macroeconomic factor affecting the value of Cambridge's housing stock in this time period, we do not interpret the evolution of d t as a causal effect of rent decontrol. 29 or r 1 and r 2 to provide unbiased estimates of the direct and indirect effects of rent decontrol on the market value of residential properties, it must be the case that the elimination of rent controls-and resulting neighborhood-level changes-must not have been fully anticipated by households and landlords. This appears plausible in light of the fact that the rent control law was narrowly eliminated ð51-49 percentÞ by a statewide referendum in which a large majority of Cambridge residents voted against rent decontrol. 30 Additionally, our identification requires that conditional on detailed geographic and time effects, the variable represent- 29 We hesitate to interpret the coefficients on the RC main effect and RCI  RC ðcoefficients l 1 and l 2 Þ as causal effects of rent control status or rent control intensity since these variables will also pick up unobserved factors that determined rent control status and rent control intensity at the time that rent control was adopted in 1970 ðe.g., the age of the residential housing stock and the fraction of nearby units that were owner-occupied vs. rentedÞ. 30 To the degree that rent decontrol and any resulting neighborhood effects were foreseen by incumbent and potential owners, buyers, and renters, these effects would substantially ing a property's exposure to rent decontrol ðRCI i  Post t Þ is uncorrelated with other unmeasured factors within neighborhoods that affect local house prices, change contemporaneously with rent control removal, but yet are not caused by the elimination of rent control. It is difficult to state precisely what these factors would be since the most obvious candidates ðe.g., improvements in neighborhoodsÞ are plausibly caused by rent control removal. We subsequently present event-study graphs with the transaction price sample that strongly suggest that the effect of rent control intensity on house prices is not present prior to the elimination of rent control and evident thereafter. In many instances, we also estimate a richer version of equation ð2Þ in which we interact the RCI measure with both the rent control main effect and the RC  Post term. This tripledifference specification allows the indirect effect of rent control intensity to differ between controlled and never-controlled properties in both the rent control and decontrol eras.

The end of rent control in 1995 coincided with a period of nationwide house price appreciation, which raises the possibility of confounding price trends. The time effects d t in our estimating model will absorb these changes to the degree that they affect the overall price level of Cambridge housing. They will not absorb any differential appreciation in rent control-intensive neighborhoods, which might hypothetically occur if, for example, the US housing boom of the early 2000s spurred an influx of lending ðand associated price appreciationÞ in rent control-intensive neighborhoods ðMian and Sufi 2009Þ. We address this concern by estimating specifications containing tract-by-year interactions, in addition to 89 geographic main effects for Cambridge block groups, thereby allowing the rate of appreciation to differ across census tracts.

A. Appreciation of Decontrolled Properties

Table 3 presents baseline estimates of equation ð2Þ for the causal effect of rent decontrol on assessed values of decontrolled properties from 1994 to 2004 using the full set of 15,475 residential properties. Column 1 reports a parsimonious specification containing only an RC main effect, an RC Â Post indicator, and a set of dummies for year of sale and structure type ðcondominium, two-family house, three-family houseÞ. Prior to rent decontrol, the assessed value of controlled ðRCÞ properties averaged 50 log points below the assessed value of never-controlled ðnon-RCÞ properties. 31 Following decontrol, this gap closed by 22 log capitalize into values before rent control was removed, which would work against our finding either a direct or an indirect effect of rent decontrol on prices. 31 The RC main effect estimates do not admit a causal interpretation, as noted above. A property's rent control status in 1994 is a function of the property's year of construction and its residential and occupancy status ðrental vs. owner-occupiedÞ as of 1971, which in turn are likely to be correlated with the fixed characteristics of the property, its mainte-points. Columns 2-4 refine the precision of the comparison by adding a set of dummy variables that sweep out cross-neighborhood differences in price levels and trends. Column 2 adds block group effects, which absorb average assessed values within narrow block groups ðaveraging 0.07 square mileÞ. Here, the model is identified by contrasting the change in market value of decontrolled and never-controlled units within block groups. Column 3 adds tract-by-year dummies, thus allowing each of the 30 census tracts in Cambridge ðaveraging 0.22 square mile in areaÞ to have a different overall appreciation rate. The final column includes a fixed effect for each residential location or map-lot ða total of 9,497 map-lotsÞ. This demanding specification, which absorbs the RC main effect, contrasts the map-lot-level change in assessed values between map-lots that contained controlled units and those that did not, again allowing for different price trends across 30 census tracts.

Across all specifications, the rent control main effect is highly robust and stable, demonstrating that decontrolled units appreciated substantially relative to never-controlled properties. These initial estimates do not distinguish, however, between the direct and indirect channels that nance and appearance, as well as the desirability of its surrounding neighborhood. While the rent control main effect is robustly large and negative in all cases, this may reflect omitted property attributes and not the causal impact of rent control.

may jointly contribute to this appreciation. In particular, since decontrolled properties are typically located in neighborhoods with aboveaverage levels of rent control intensity ðtable 2Þ, the RC Â Post term estimated above captures a combination of direct decontrol effects and indirect ðmicro neighborhood levelÞ effects stemming from the greater desirability of formerly rent control-intensive locations. Our next set of estimates distinguishes these two effects.

B. Direct and Indirect Effects of Rent Decontrol

Table 4 augments the simple difference-in-difference models above with a measure of the rent control exposure of each residential property ðdenoted RCI and calculated using a 0.20-mile radiusÞ, as well as an interaction term between the RCI measure and a post-1994 indicator variable.

This term measures the degree to which properties with greater rent control exposure saw differential appreciation following decontrol. The inclusion of the RCI and RCI Â Post measures also changes the interpretation of the RC Â Post main effect. Whereas previously this variable measured the differential appreciation of decontrolled versus nevercontrolled properties averaging ðimplicitlyÞ across more and less rent control-intensive areas, the RC Â Post coefficient in the augmented specification measures the differential appreciation of RC relative to non-RC properties in a hypothetical location with no other surrounding controlled properties ði.e., RCI 5 0Þ. The base specification in column 1, which contains only year of sale and structure type dummies in addition to the RC and RCI terms, finds that properties with higher rent control exposure had lower value in the decontrol era and that this differential was substantially reduced in the period following decontrol. Specifically, the point estimate of 20.58 on the RCI measure indicates that a property at the mean level of rent control exposure of 0.32 was assessed at approximately 19 log points below a property with zero exposure. We do not take the main effect of the RCI variable to be causal, however, since it is likely to be correlated with the many factors that determined which properties were controlled in 1971. Conversely, the coefficient of 0.33 on the RCI Â Post indicator implies that 56 percent of this price differential was erased in the years after decontrol. Under our identifying assumption that these unobserved factors are quasi-fixed or are not spuriously correlated with rent control intensity across local areas, the RCI Â Post interaction may be viewed as a causal estimate of the indirect effects of rent control on the market value of surrounding units ðboth formerly and never controlledÞ. The fact that both the RC main effect and the RCI Â Post coefficients fall in magnitude relative to the table 3 estimates ðwhich exclude the RCI measureÞ reveals that the lower market value of RC properties stems in part from the fact that they were situated in more rent control-intensive locations. We explore the robustness of these initial relationships by applying the control variables used above: block group fixed effects, tract-year effects, and map-lot fixed effects. While these covariates reduce the precision of the RCI main effect, the point estimates remain large and statistically significant. The coefficient of primary interest ðRCI Â PostÞ increases in magnitude with the inclusion of tract-year effects. Column 3 obtains an estimate for the RCI Â Post coefficient of approximately 55 log points, while in column 4 the estimate is 48 log points, implying that a residential property at the 75th percentile of rent control exposure gained approximately 13 percent more in assessed value following decontrol than a property at the 25th percentile of exposure. 32 hese first four estimates constrain the indirect effects of rent control to be identical for never-controlled and decontrolled units. In practice, these effects may differ. Indeed, if the indirect effect were present only for decontrolled units, this would suggest that the indirect effect is not operating through the hypothesized localized amenity channel ðwhich we would expect to affect both property typesÞ. The models in columns 5-7 demonstrate that both decontrolled and never-controlled properties benefit from the indirect effect of rent control removal. In the most demanding specification in column 7, which includes map-lot fixed effects and tract-by-year dummies, we estimate an RCI Â Post coefficient of 0.42 for never-controlled properties and 0.61 for decontrolled properties. Both are significantly different from zero, and the data do not reject the hypothesis that these coefficients are of the same magnitude.

Notably, the models that allow for separate indirect effects for controlled and never-controlled properties also find that the RC discount is only approximately half as large as was implied by the earlier models that do not include interaction terms between RCI and rent control statusapproximately 23 rather than 48 log points-and that this RC discount was fully offset by the post-decontrol appreciation of decontrolled properties. By implication, approximately half of the estimated RC discount detected in columns 1-3 of the table is accounted for by the fact that RC units were situated in more rent control-intensive locations and that rent control exposure differentially lowered their value. While our conceptual model is silent on why the indirect effect of rent control is greater for controlled than for never-controlled properties, one speculative explanation is that deferred maintenance and poor property management were more acute in locations where a larger fraction of properties were controlled. This conjecture would also be consistent with our finding of greater relative appreciation of decontrolled than never-controlled units in rent control-intensive locations, though, as above, this differential is not statistically significant.

C. Variation across Property Types

Table 5 explores the potentially differing consequences of rent decontrol for the assessed values of houses and condominiums. 33 Across the two panels, the direct impact of rent decontrol on controlled houses is 32 The 25th and 75th percentiles of the RCI distribution are 0.464 and 0.199. The implied interquartile effect is 0.126 5 0.475 Â ð0.464 2 0.199Þ. 33 To simplify exposition, we display only the interaction terms between post-decontrol and the RC and RCI terms, suppressing the included main effects of these variables. substantially smaller than the corresponding estimate for condominiums, a pattern that may be due to the greater extent of upgrading at controlled condominiums. 34 For residential houses ðpanel AÞ, we estimate an indirect effect coefficient on the value of residential houses of approximately 20 log points, implying that a house facing the mean level of RCI of 0.37 would experience an additional 7.4 log points of appreciation relative to a nonexposed house following rent decontrol. When allowing for separate indirect effects for decontrolled and never-controlled houses ðcols. 3 and 4Þ, we find that the indirect effect for decontrolled houses is 20-50 percent larger than for never-controlled houses, although we are unable to reject the equality of the two regression coefficients. Adding 9,497 map-lot fixed effects to the regression model in column 4 decreases precision such that the indirect effect estimates for houses become insignificant, though magnitudes are only modestly affected.

The parallel analysis for condominiums in panel B finds significant indirect effects of rent decontrol on both decontrolled and nevercontrolled condominiums. The indirect effects for condominiums are greater than for houses, although standard errors are also considerably larger. The point estimate of 0.49 in column 2 implies differential appreciation of 18.2 log points for a condominium at the mean level of rent control exposure relative to a nonexposed unit. As is the case with houses, adding a map-lot fixed effect for each land parcel containing condominiums ð1,450 fixed effectsÞ reduces or eliminates the statistical significance of the indirect effect point estimates.

The number of condominiums in Cambridge rose by one-third between 1994 and 2004, with almost half of this rise due to the conversion of existing houses to condominiums. This substantial change in the housing stock implies that part of the rise in assessed values may be due to capital improvements in residential units, particularly condominiums, rather than solely to changes in the value of ownership stemming from decontrol ðe.g., the option to charge higher rents or convert the unit to owner-occupied statusÞ. Notably, this concern applies only to the direct effects estimates of rent decontrol on decontrolled units, which may conceivably combine both the investment and ownership channels. For the indirect effects we measure, this source of variation-spillovers from local housing investments spurred by rent decontrol-is not a concern; indeed, this is one of the key causal channels through which we hypothesize the indirect effects operate. 35 34 As discussed in Sec. VIII, Cambridge building permit data indicate that annual citywide investments in decontrolled condominiums increased by 206 percent in the post-vs. pre-decontrol period while the corresponding increase for decontrolled houses was 120 percent ðtable A1Þ. 35 The converse concern applies, however: if localized spillovers spur additional investments at decontrolled units, this may contribute to the estimated direct and indirect To explore the importance of the investment channel, we undertake two exercises. We first reestimate the main models for direct and indirect effects of decontrol on houses and condominiums while excluding all units that changed usage categories ðe.g., converted condominiumsÞ between 1994 and 2004. Then, in Section VII, we directly explore the role of investments using Cambridge residential building permit data.

When dropping converted properties from the regression estimates in columns 5 and 6 of table 5, we find little effect on the direct or indirect effect point estimates for houses. This is sensible since only one in 10 houses changed status during the 10-year window. Consistent with our reasoning above, however, the estimated direct effect of rent decontrol on condominium values is substantially reduced when converted properties are excluded-falling by as much as 35 percent between columns 3 and 5-while the indirect effects estimates are substantively unaffected. 36 hese results suggest that the direct effects estimates capture both capital improvements and changes in ownership value spurred by decontrol, particularly for condominiums. The indirect effects estimates, however, are not affected ðconceptually or empiricallyÞ by abstracting from the substantial investments made in converted units.

D. Testing Alternative Measures of Rent Control Intensity

Our estimates so far employ a measure of rent control intensity calculated over a 0.20-mile radius from each Cambridge map-lot. We explore the sensitivity of the results to this choice by employing two sets of alternative measures: one that varies the geography over which rent control intensity is calculated and one that varies how the weight given to surrounding properties decays with distance. These robustness tests employ the final ðmost exhaustiveÞ specification in table 4, which includes fixed effects for each individual map-lot and tract-by-year dummies that account for differing rates of property appreciation across all 30 Cambridge census tracts.

Panel A of table 6 reports estimates using RCI measures calculated at differing geographies. Columns 1-3 calculate RCI at radii of 0.10, 0.20, and 0.30 mile, respectively ðhence, col. 2 replicates our main specification from table 4Þ. Column 4 instead uses an RCI measure calculated at the level of 587 census blocks. Distinct from the radius-based measures, census blocks comprise a set of contiguous, nonoverlapping geographic subdivisions. These blocks may not, however, correspond to any specific notion of neighborhood or proximity, particularly since residential units on opposite sides of the same street are often assigned to different blocks. The direct impact of decontrol is relatively insensitive to the geography of the RCI measure. Across specifications, this effect averages 13-15 log points, which is generally not significant. 37 The indirect effect estimates are somewhat more sensitive. The largest indirect effect estimate comes from the 0.30-mile radial RCI measure, though this effect is statistically indistinguishable from the 0.20-mile measure. Estimates using the census block group-based RCI measure are generally insignificant, which is consistent with our observation that block boundaries do a poor job of capturing neighborhood proximity.

An important conceptual limitation of the radius-based measure is that it puts equal weight on each residential unit within a specified radius of a given map-lot while simultaneously according zero weight to all other units in Cambridge. It seems plausible, however, that the interactions among residential units decline with distance, so that nearby units matter more for a unit's rent control intensity as perceived by occupants and potential buyers, while more distant units matter less.

To explore this idea empirically, we employ an alternative measure of rent control intensity that places greatest weight on nearby units and less weight on more distant units. Specifically, we use an exponential decay function to calculate the RCI of each unit i as a function of its distance from all other RC ðrent-controlledÞ units j ≠ i in Cambridge, where the weight given to each unit j is declining in its distance from i. Let d ij be the distance between units i and j measured in miles and l < 0 be a negative constant, J be the complete set of residential units in Cambridge, and RC j be a dummy variable equal to one if unit j is rent controlled and zero otherwise. Our distance-based measure of RCI l is

Like the primary RCI measure, the measure RCI l i lies on the unit interval. The difference between RCI l i and RCI is that the weight given to surrounding units in RCI l i is a continuous, declining function of distance from i whereas, for RCI, the weighting function is flat over the area of the designated radius and then is equal to zero outside of that area.

Panel B of table 6 reports estimates of this decay-based RCI measure using values of l ranging from 212 to 23, where lower ðmore negativeÞ values of l give greater weight to nearby units and higher values of l give greater weight to more distant units. To illustrate the operation of our weighting function, the first several rows of the panel display the weight accorded to units at 0.10, 0.20, and 0.30 mile from each reference unit relative to units at 0.01 in the RCI calculation. All four values of l accord a weight that is close to unity to properties within a radius of 0.01 mile and under, whereas more negative values of l place substantially less weight on distant units. 38 For example, at l 5 212, units at 0.20 and 0.30 mile receive weights of 0.10 and 0.03 relative to the weight at 0.01 mile, respectively. In contrast, with l 5 23, these units receive relative weights of 0.57 and 0.42, which are substantially greater. The estimated direct and indirect effects of rent decontrol on residential property appreciation are both stable and robust across the four parameterizations of the decay function. In all cases, the indirect effects estimates are statistically significant for both decontrolled and never-controlled properties and are comparable in magnitude to the radius-based measures, though, if anything, the decay-based estimates are more robust.

One pattern evident in both panels of table 6 is that the estimated magnitude of the indirect effect rises when we use an RCI measure that gives greater weight to more distant properties ðby employing a wider radius or a more gradual decay functionÞ. A likely explanation for this pattern is that employing a broader RCI measure provides more information about the extent of a unit's rent control exposure; that is, a given high ðor lowÞ RCI value obtained over a larger radius ðor slower decay functionÞ implies that the relevant unit is more ðor lessÞ deeply surrounded by other controlled units. This should in turn imply a larger indirect effect of RCI on market values. Logically, the variance of the RCI measure declines as its scope broadens as shown in the lower row of each panel, so the size of the standardized effect rises less rapidly than do the RCI point estimates. 39 longside the robustness tests in table 6, in online Appendix B, we have explored a variety of alternative and complementary identification strategies that probe the key results. One potential limitation of our primary approach stems from the nonparallelism between the RCI measures and the geographic dummy variables used as controls. While the radius-based RCI measure in many cases partially overlaps multiple block groups, each map-lot in the regression is associated with only a single block group fixed effect. To explore sensitivity to this choice, we created a set of "rolling" block fixed effects. For each map-lot, we identify each census block ðof which there are 587Þ whose centroid lies within 0.2 mile of the map-lot and assign these block dummies to the map-lot. These non-mutually exclusive block fixed effects are then used in place of the conventional block group fixed effects in the regressions ðsee table B1Þ . We perform an analogous exercise for the exponential decay specifications, where all block dummies in Cambridge are fractionally assigned ðsumming to unityÞ to each map-lot as a decaying function of the distance between the map-lot and block centroids ðsee table B1Þ . Motivated by the fact that our radius-based rent control intensity cannot account for the characteristics of non-Cambridge properties bordering the city, we perform a third specification test that obviates this issue by excluding all Cambridge block groups that border non-Cambridge properties ðsee table B2Þ . All three of these sensitivity exercises, reported in Appendix B, yield estimates that are highly comparable in magnitude and precision to our primary estimates above.

VI. The Time Path of Rent Decontrol Capitalization

Because our assessor data cover only two points in time, 1994 and 2004, they do not shed light on how residential real estate prices evolved prior to decontrol or in the years thereafter. We turn to housing transaction data to complete this picture. We begin with simple "event-study" plots of the main effect of rent control status on real estate transactions, estimated with the equation

where Y S igt is the real sales price of residential unit i located in block group g in year t, the vectors g g and d t contain fixed effects for block groups and year of sale, and the X vector contains a rich set of property characteristics, sourced from deed records and summarized in Appendix table A4, including the count of rooms, bathrooms, and bedrooms; the unit's interior square footage; a quadratic in lot size and a dummy for lot size equal to zero ðcommonplace for condominiumsÞ; and a quadratic in the log of property age and a dummy for missing year built. All controls are interacted with dummies for structure type ðcondominium, singlefamily home, multifamily homeÞ since the hedonic value of these attributes may differ across types, and structure type dummies are further interacted with quadratic time trends to allow for differing price trends. Standard errors are clustered at the block group level. Figure 3 plots the key coefficients ðr 1;t Þ from equation ð4Þ, which correspond to by-year estimates of the rent control price differential measured relative to the omitted reference year of 1994.

The relative price of RC properties increased by roughly 10 log points over the first 3 years following decontrol, declined very modestly between years 3 and 4, and then rose almost continuously thereafter. By the end of the sample in 2005, RC properties had increased in market value by almost 30 log points relative to nearby non-RC properties with similar characteristics. The increasing cumulative effect of decontrol on transaction prices parallels the evidence above on the evolution of resident turnover, which also rose immediately following decontrol and then generally trended upward through the end of the sample window. Both results suggest that changes in the desirability of locations and neighborhoods induced by decontrol likely took years to unfold.

We plot the indirect effects of rent decontrol on the value of nevercontrolled and decontrolled properties in figure 4 using a specification analogous to equation ð4Þ augmented with RCI Â Year terms. 40 Indirect price effects of decontrol on the sale prices of never-controlled properties begin to accumulate immediately following decontrol, attain statistical significance by the fifth year following decontrol ð1999Þ, and 40 The specification is estimated separately for never-controlled and decontrolled units, and hence RC Â Year effects are not included. continue to rise through the end of the sample, yielding a point estimate of 0.60 log points in 2005 ðp < .01Þ. Indirect effects estimates for decontrolled properties ðpanel IIÞ offer a similar picture of post-decontrol appreciation: the indirect effect for decontrolled units varies in sign and is generally insignificant in the pre-decontrol years; following decontrol, the point estimates are strongly positive in 10 of 11 years and are statistically significant at p < .10 in 7 of 11 years. Though the event-study plot is substantially noisier for decontrolled than for never-controlled properties, this likely reflects the fact that there are only half as many transactions of decontrolled units between 1988 and 2005 ð4,802 relative to 9,987Þ.

Table 7 explores these relationships in further detail. Columns 1-3 provide estimates of the direct effect of decontrol on the market value of formerly controlled properties. The estimates range from 6 to 11 log points as we add block group fixed effects and controls for property characteristics ðcol. 2Þ. 41 Taking advantage of the additional years of data available from the transactions sample, column 3 adds linear and quadratic trends for each of the 30 census tracts ðalso interacted with property typesÞ to allow for the flexible evolution of real estate prices over time within fine geographies. These controls have little impact on magnitudes or precision. Subsequent columns introduce the RCI measure to estimate indirect effects. The coefficient of interest in column 4 ðRCI Â PostÞ is 21 log points, which is in the lower range of estimates obtained using the assessor's sample ðtable 4Þ. As figure 3 suggests, however, both the direct and indirect effects of rent decontrol cumulate over the sample window, so that the average post-decontrol effect is smaller than the long-run effect estimated by the long difference specifications used with the assessor data. Adding flexible geographic trends ðcol. 5Þ slightly reduces the indirect effect estimate and increases its standard error.

Columns 6 and 7 report estimates in which the indirect effect of rent control exposure on prices is permitted to differ between decontrolled and never-controlled properties. The indirect effect estimates are statistically indistinguishable between decontrolled and never-controlled units, though they are generally larger for decontrolled properties and more precisely estimated for never-controlled properties ðconsistent with fig. 4Þ. In column 7, which includes all prior covariates plus tractspecific quadratic trends, the point estimates for the indirect effects are Note.-N 5 14,789 Cambridge house and condominium properties transacted during 1988-2005. Prices are Winsorized by structure type at the 1st percentile. RCI is defined over a 0.20-mile radius and demeaned. RC is an indicator for a location that was rent controlled in 1994, and Post is an indicator for year 1995 or afterward. All specifications include year of sale dummies and structure type dummies. Property characteristics, each interacted with structure type, include number of total rooms, bathrooms, bedrooms, interior square footage, a dummy variable for zero lot size, a quadratic in lot size, a dummy variable for missing year built, a quadratic in the log age of the structure, and a quadratic time trend for each structure type. Block group fixed effects correspond to each of the 88 Cambridge block groups in the 1990 census containing transacted properties. no longer significant at conventional levels, though the point estimates are little affected.

Echoing the results from property assessments, the transaction data models suggest that the price penalty for rent control exposure was substantially greater for controlled than for never-controlled houses in the rent control era. Accounting for this differential in columns 5-7 reduces the magnitude of the estimated RC price discount prior to decontrol. For example, the RC point estimate of 217 log points in column 6 is quite comparable to the RC point estimates in table 4 that use property assessments and allow for separate RCI slopes for RC and non-RC units. While the direct effect of decontrol on RC units appears modestly smaller in magnitude in the transaction-based models than in the assessment-based models, this again likely reflects timing, with the assessment-based models using a 10-year-long change and the transaction-based models using a series of annual observations. The transaction sample also provides an opportunity to explore the concern that assessed real estate values may not accurately reflect market prices. We examined this issue by constructing a matched assessor's sample for properties that transacted in 1994 and 2004, and we used this matched set to perform parallel estimates of the direct and indirect effects of decontrol on assessed and transacted property values in Appendix table A5 . These models yield a close match between the estimated decontrol impacts on assessed values and transaction prices, and moreover, the match is particularly close for houses. 42 This comparison suggests that the transaction and the assessor's sample provide complementary and broadly consistent measures of valuations.

We also explored the possibility that rent decontrol had positive indirect effects on the market value of properties close by Cambridge, focusing on the adjoining town of Somerville. For this exercise, we assembled residential transaction records for Somerville analogous to those used above for Cambridge, limiting the sample to residential units located in census tracts and block groups that abut either the Somerville-Cambridge border ðnorth of CambridgeÞ or the Somerville-Medford border ðnorth of SomervilleÞ. 43 Comparing properties along these two Somerville borders, in Appendix table A6 we find robust evidence that 42 An additional complexity in comparing the assessed vs. sale values of condominiums is that we are unable to determine which specific unit among the assessed condominium units at a map-lot is transacted. Consequently, we include matched assessor data for all units at the map-lot in which one or more units transact. This leads to a sample of 7,897 condominium assessments matched to the 937 units that were transacted in 1994 or 2004.

43 Somerville is bordered by Cambridge, Arlington, Medford, Everett, and Charlestown. Its two longest borders by a considerable stretch are those with Cambridge in the south and Medford in the north. The transaction data we assemble are also sourced from the Warren Group files, used for the price analysis immediately above, and contain the identical data elements and years of coverage.

Cambridge-bordering properties appreciated by 7-10 percent more than Medford-bordering properties with the same observable characteristics. This pattern is consistent with the hypothesis that improvements in Cambridge neighborhood amenities following decontrol also increased the desirability of locations bordering Cambridge.

Online Appendix B reports on additional robustness tests that investigate price appreciation in neighboring towns, explore the potential importance of the increase in subprime credit in lower-income Cambridge neighborhoods during this time period, and test for correlations between rent control intensity and changes in the composition of Cambridge properties that transact. These many specification tests, in combination with our prior estimates using assessed property values, confirm that alongside its direct effects on the market value of formerly controlled units, rent decontrol had robust indirect impacts on market values. Both decontrolled and never-controlled properties in Cambridge that were more exposed to controlled units saw differential appreciation in the post-decontrol regime. Before benchmarking the economic magnitudes of these direct and indirect effects, we briefly consider one plausible channel through which they may have operated: property investments.

VII. The Impact of Rent Decontrol on Property Investments

Cambridge experienced an overall investment boom after the end of rent control. Total permitted investment at houses and condominiums rose from $83 million in the period 1991-94 to $455 million in the period 1995-2004, while annual investment expenditures roughly doubled at three of four property types-decontrolled houses, never-controlled houses, and never-controlled condominiums-and roughly tripled at decontrolled condominiums ðtable A1Þ. 44 While fewer than one in 25 residential units receives a building permit annually, this fraction increased substantially following decontrol: by 17 percent and 7 percent among never-controlled and decontrolled houses and by 38 percent and 45 percent among never-controlled and decontrolled condominiums. 45 Was this increase in residential investment caused by rent decontrol? 44 As detailed in App. B, our analysis in this section draws on a database of all building permits issued by the Cambridge Inspectional Services Department for years 1991-2005, including property address and proposed expenditure. Since permits can be filed either for a structure ðe.g., a multi-unit condominium complexÞ or for any unit in a structure, we attribute a permit at a given structure to only one unit in that structure when computing permitted units in table A1 .

An event-study of the direct impact of decontrol on permitting activity ðfig. 5Þ shows a sharp, statistically significant differential rise in permitting activity and investments at decontrolled relative to never-controlled properties during the first 5 years following decontrol. 46 Relative to never-controlled units, the annual permitting probability at decontrolled units rose by 2.5-3.0 percentage points while permitted investment expenditures per decontrolled unit rose by $1,000-$1,500. We find no evidence, however, that decontrol indirectly caused permitting or investments to rise in decontrolled or never-controlled units with higher rent control exposure ðnot shownÞ. Thus, any indirect effect of decontrol on residential investments that took place was not localized to highly rent control-exposed units, though it remains plausible that decontrol helped to spur the citywide investment boom documented in table A1 . One factor that may obscure any direct expenditure effect in our analysis is low statistical power: the vast majority of investment expenditures are zero, while the mean and variance of expenditures at permitted units are high and rising ðtable A1, panel CÞ. However, unless increased investment occurred along dimensions that do not require permits and hence are not observed in our data-for example, repairs and maintenance that are not structural and do not alter major systems-the pattern of results appears to rule out very large differential expenditure effects at formerly controlled units. We consider an upper bound on the contribution of induced investments to post-decontrol property appreciation in the next section.

VIII. The Capitalized Value of Rent Decontrol in Cambridge

How economically large are the direct and indirect effects of rent control estimated above? We answer this question by benchmarking our estimates against the overall level of house price appreciation in Cambridge using the Cambridge Assessor's Database as our measure of the value of the housing stock. Panel A of table 8 presents information on the assessed value of the Cambridge housing stock in 1994 and 2004. Between 1994 and 2004, the assessed value of houses and condominiums rose from $4.7 billion to $12.5 billion in constant 2008 dollars, a gain of 163 percent. Notably, appreciation of decontrolled units exceeded that of never-controlled units by a substantial margin-219 percent relative to 152 percent-and these gains were larger among both decontrolled houses ð185 percent vs. 147 percentÞ and decontrolled condominiums ð237 percent vs. 166 percentÞ.

What was the contribution of rent decontrol to these gains? We compute the direct and indirect contributions of rent decontrol to these valuations by applying our most conservative regression estimate, which includes map-lot fixed effects and census tract trends ðtable 4, col. 7Þ, to calculate the counterfactual price change at each location between 1994 and 2004, assuming that rent control had remained in place. While the aggregate value of decontrolled houses and condominiums increased from $785 million to $2.5 billion between 1994 and 2004, our estimates imply that had rent control not been eliminated, this gain would have been $849 million smaller, with $310 million due to the ðforegoneÞ direct effect of decontrol and an additional $539 million due to the ðforegoneÞ indirect effects on decontrolled properties. By implication, about half of the $1.7 billion appreciation of decontrolled properties between 1994 and 2004 can be accounted for by rent decontrol.

Though never-controlled units do not ðby definitionÞ benefit directly from decontrol, the indirect effects of decontrol are substantial. Our estimates imply that of the $6.0 billion gain in the assessed value of never-controlled houses and condominiums between 1994 and 2004, $1.1 billion ð13 percentÞ of this gain is due to the indirect effects of rent decontrol on never-controlled properties, with $822 million accruing to houses and $306 million to condominiums.

Putting these components together, we calculate that decontrol added almost exactly $2.0 billion to the value of the Cambridge housing stock between 1994 and 2004, with 84 percent of this effect due to the indirect effect of rent decontrol. While the share of post-decontrol appreciation between 1994 and 2004 induced by rent control removal was substantially larger for decontrolled than for never-controlled properties ð49 percent vs. 19 percentÞ, the never-controlled segment of the market received the largest increase in capitalization from rent control's removal: $1.1 bil- lion versus $849 million. By implication, prior to decontrol, the nevercontrolled sector bore more than half of the incidence of rent control regulation.

Can the increase in residential investments documented in Section VII account for these price impacts? Total permitted residential investments averaged $45.5 million between 1995 and 2004 ðtable A1Þ. 47 In the 4 years prior to decontrol, these expenditures averaged $20.8 million. To benchmark the maximal estimate for the capitalized value of these investments, consider a case in which the entire $24.7 million increase in annual expenditures could be causally attributed to rent decontrol, where each dollar of expenditure led to a dollar of price appreciation and where there was no subsequent depreciation of these investments during this 10-year interval. In this case, we would conclude that only 12 percent of the appreciation of Cambridge residential properties between 1994 and 2004 was due to increased investments induced by rent decontrol ð$247 million of $1,977 billionÞ, leaving the remaining 88 percent accounted for by the capitalization of other benefits of rent decontrol. A parallel set of calculations implies that increased investments can explain at most 18 percent of the indirect effect of decontrol on the value of never-controlled properties and at most 6 percent of the total ðdirect plus indirectÞ effect of decontrol on the value of formerly controlled properties. In fact, our event-study estimates in figure 5 imply that decontrol can account for no more than $82 million of the total increase in Cambridge residential investments in the years following decontrol, all of it concentrated on decontrolled properties. 48

IX. Conclusion

The largely unanticipated elimination of rent control in Cambridge, Massachusetts, in 1995 affords a unique opportunity to identify spillovers in residential housing markets. This paper exploits the sharp crossneighborhood contrasts in the fraction of units that were decontrolled to assess the localized price spillovers to never-controlled properties as well as to quantify direct effects on decontrolled properties. Our main finding is a large and significant positive indirect effect of decontrol on the valuation of properties that were exposed to controlled units, leading, on average, to a 16 percent increase in the value of residential units between 1994 and 2004. We further document that rent-controlled properties were valued at a substantial discount relative to nevercontrolled properties and that rent decontrol eliminated a substantial part of this differential, raising the assessed values of these properties by approximately 13 -25 percent.

The contribution of decontrol to the capitalized value of the Cambridge residential housing stock in this period corresponds to a total of $2.0 billion. While the direct effects on decontrolled properties were larger in percentage terms than the effects on never-controlled properties, the stock of controlled properties was smaller and less valuable than the never-controlled stock. As a consequence, indirect effects on never-controlled properties account for more than half ð56 percentÞ of the decontrol-induced increase in the value of the housing stock.

Because, under any reasonable set of assumptions, increases in residential investment stimulated by rent decontrol can explain only a small fraction of these indirect effects, we conclude that decontrol led to changes in the attributes of Cambridge residents and the production of other localized amenities that made Cambridge a more desirable place to live. This possibility is also highlighted by our theoretical model, though we are not able to thoroughly examine it with our data. Glaeser and Luttmer ð2003Þ argue that nonprice rationing under rent control leads to a mismatch between renters and apartments and provide evidence that this allocative inefficiency is large in New York City's rent control plan. It is therefore reasonable to conjecture that the unwinding of allocative distortions significantly contributed to Cambridge's residential price appreciation. Additional empirical analysis with rich microlevel attributes of residents, however, will be needed to shed further light on rent control's allocative consequences.

A key issue in the evaluation of price controls is the trade-off between the surplus transferred from landlords to renters and the deadweight loss from quality or quantity undersupply. Viewed in this light, some portion of the price gains we measure at decontrolled properties are transfers from renters back to landlords. However, our analysis highlights the importance of another welfare consequence of price controls: the indirect effect on the desirability of housing in rent controlintensive locations. Our results indicate that the efficiency cost of Cambridge's rent control policy was large relative to the size of the transfer to renters. In particular, only 16 percent of the capitalized value of rent decontrol reflects a direct impact on decontrolled units, with the remainder due to indirect effects of rent control exposure on the amenity value of Cambridge residential units.

These findings are germane to economic analysis of housing market regulations and, more broadly, to the impacts of other place-based policies. The mechanisms by which rent decontrol affects never-controlled housing-increased maintenance, upgrading of local amenities, and potentially more efficient sorting of consumers to housing-are likely present in other settings involving residential housing. Our results provide evidence that residential spillovers are large and important in housing markets and suggest that public policies related to housing should consider not only direct impacts but also indirect impacts on neighboring properties and residents.

Appendix A Theory

We ground our empirical analysis in a stylized equilibrium model of the housing market that considers the relationship between rent control, neighborhood amenities, and house prices.

Neighborhoods.-A city consists of n 5 1, . . . , N neighborhoods. There is a continuum of locations in each neighborhood indexed by ' ∈ ½0; 1. The pair ð,, nÞ refers to location , in neighborhood n.

Landlords.-Each location is owned by an absentee landlord who decides on the level of maintenance m. Maintenance includes inputs such as painting, upgrading, and repairs. These produce housing services according to the following increasing and concave technology: h 5 f ðmÞ. While the model is static, we interpret housing services as a per-period flow variable. The price of housing services, p, is a per-period price.

The cost of maintenance is given by an increasing and convex function c ðmÞ. The problem of the landlord is to choose a maintenance level m to maximize profits: max m ph 2 cðmÞ:

The first-order condition for an interior solution implies that maintenance is an increasing function of the price of housing services. Denote this function as m * 5 mð pÞ, where m 0 ð pÞ > 0.

Residents.-Residents have preferences given by U ðc; hÞ 5 Ac 12a h a ;

where c is a composite commodity, h is housing services, and A is the total level of amenities in the neighborhood. The price of housing at location , in neighborhood n is denoted p n ð,Þ, so a resident who lives at ð,, nÞ faces the budget constraint

where y denotes income. The only heterogeneity in the model comes from differences in income y between residents. The outside utility for a resident with income y is denoted by U y .

Amenities depend on neighborhood attributes. To capture the most relevant dimensions for our study, we assume that amenities are increasing in the overall level of maintenance and income of residents as follows:

Here, m n ð,Þ denotes the maintenance level at location , in neighborhood n and y n ð,Þ denotes the income of residents in neighborhood n residing at location , and b ∈ ½0; aÞ. The equilibrium concept is based on spatial equilibrium, with free entry and perfect mobility of residents.

Equilibrium definition.-An equilibrium is a triple h y n ð'Þ; p n ð'Þ; h n ð'Þi where y n ð,Þ is the resident income, p n ð,Þ is the price, and h n ð,Þ is the level of housing services for each neighborhood n and location , such that

• each resident obtains at least his outside option,

• no resident wishes to move to another neighborhood or location within a neighborhood, and

• landlords maximize profits.

Benchmark model.-We impose particular functional forms to keep the model tractable. For the supply side, assume that housing services are produced by the linear technology f ðmÞ 5 m and the costs of maintenance are quadratic: cðmÞ 5 ð1=2Þm 2 . These assumptions imply that the optimal level of maintenance at each location is exactly equal to the price of housing services: m * 5 p. The demand for housing services decreases with price: h 5 ay=p. Next, we assume that the distribution of income among potential residents consists of N distinct levels of y, which we order from highest to lowest, y 1 > Á Á Á > y N .

We first solve for the equilibrium without rent control as a baseline. We then consider the controlled equilibrium and develop implications for how prices, maintenance, and resident allocation will be affected by decontrol.

Equilibrium without Rent Control

We consider a symmetric equilibrium in which all residents with income y n live in neighborhood n. The log indirect utility V for a resident of neighborhood n at location , is ln

Free entry and perfect mobility of residents imply that in all locations , in neighborhood n, each resident's utility is equal to U yn . Hence, the price of housing services at each location , is

The value of neighborhood amenities comes from the fact that landlords optimally set the level of maintenance to the price of housing services and in the candidate equilibrium all residents of neighborhood n have income y n . Therefore, ln ðp n ð'ÞÞ 5

Symmetry among landlords implies that maintenance levels within a neighborhood are the same at each location, so that

This relationship captures the feedback between overall maintenance in the neighborhood and location-specific maintenance choices. The maintenance levels in the uncontrolled economy m u n are

and prices are identical at all locations , within neighborhood n. From the expression for the level of maintenance, the price of housing p u n in neighborhood n in the economy without rent control is

The pricing equation illustrates intuitive patterns under our parameter assumptions ð1 > a > b ≥ 0Þ. Prices are higher in neighborhoods when residents have more income and they are lower when residents have better outside options. Landlords invest more in response to exogenous improvements in neighborhood quality because more investment in the neighborhood raises amenities, which raises prices, and landlords set maintenance in response to prices.

Equilibrium with Rent Control

Let RC n ⊂ ½0; 1 denote the set of rent-controlled locations in neighborhood n. Suppose that a fraction l n of locations are rent controlled and 1 2 l n are not. We first examine the pricing and maintenance decisions at controlled locations.

Rent-controlled locations.-Suppose that the rent control authority sets prices at controlled locations p n ð'Þ, and we assume that for each controlled location, the controlled price is less than the corresponding price in the uncontrolled economy, p n ð'Þ < p u n . This price will determine the level of maintenance according to the producer's first-order condition, which yields m n ð'Þ 5 p n ð'Þ:

In turn, the amount of housing services at location , is given by

Uncontrolled locations.-Spatial arbitrage determines the prices of uncontrolled locations, and hence, the arbitrage relation in equation ðA1Þ determines prices. Since m n ð'Þ 5 p n ð'Þ and landlords are symmetric at uncontrolled locations, the level of amenities in the controlled economy is

As with the uncontrolled economy, we focus on the equilibrium in which y n ð'Þ 5 y n for all uncontrolled locations ,. This yields

Since it is set by the rent control authority, the price of all controlled locations in neighborhood n may differ at each location, so we cannot further simplify the first term. For controlled locations, the income of a resident y n ð,Þ depends on the way in which residents are assigned to controlled housing. Let

While we do not explicitly model how residents are assigned to controlled housing, we assume that

which implies that the rationing mechanism imposed by rent control yields misallocation relative to the equilibrium in the uncontrolled economy. 49 The basis for this assumption is the following. If, prior to the implementation of rent control, the allocation were as in the symmetric equilibrium without rent control above, then once rent control is implemented, maintenance levels and hence housing services fall at controlled units ðsince landlords choose maintenance levels facing a regulated priceÞ. The combination of reduced rents and lower maintenance has one of two effects on incumbent residents: either they are sufficiently compensated by reduced rents so that they remain at their current locations, although the bundle of maintenance and amenities is not optimized for their income; or, alternatively, they choose to relocate to areas with higher amenities and higher rents. In the latter case, they will be replaced by residents who prefer lower housing services, that is, those with lower income levels. The average income at controlled locations will therefore weakly decline following the imposition of rent control. As a result, amenities in neighborhood n in the presence of rent control are given by ln

To compute the level of maintenance in uncontrolled locations in the presence of rent control, we follow similar steps to find

We can write this in terms of the level of maintenance at uncontrolled locations in the economy without rent control:

In summary, since neighborhood amenities are a function of the maintenance of all units in a neighborhood and the income of residents, the supply of amenities at noncontrolled locations in neighborhoods with rent controls-as well as maintenance and rents-is also impaired by rent control. This causes lower-income residents to move into noncontrolled locations. Hence, imposition of rent control causes inefficiently low maintenance and misallocation of residents at both controlled and noncontrolled locations within a neighborhood.

The Effect of Rent Control Removal on Rents, Maintenance, and Resident Allocation

Consider finally the impact of rent decontrol on prices at uncontrolled locations.

To form this comparative static, we compare price levels in the economy without and with rent control:

Dlnðp n ð'ÞÞ 5 lnðp u n ð'ÞÞ 2 lnðp c n ð'ÞÞ 5

n pertains to the maintenance effect being greater than zero, and lny n 2 k 2 n pertains to the allocative effect being greater than zero.

This expression shows that the end of rent control generates price impacts on uncontrolled locations through two channels in the model. For a given neighborhood, under rent control, maintenance is inefficiently low and there are allocative inefficiencies due to the assignment of residents at controlled locations. This expression also illustrates three natural comparative statics. When a neighborhood has a higher fraction of locations that are controlled ðl n in-creasesÞ, the change in prices for locations without rent control increases. As k 1 n increases ðas would be expected when the prices of controlled locations are further depressed from their market valuesÞ, the change in the price of uncontrolled locations due to the elimination of rent control also increases. Moreover, when there is greater misallocation due to the rent control ðk 2 n decreasesÞ, the elimination of rent control further increases prices.

The price impact due to the end of rent control for formerly controlled locations involves an additional term that can be decomposed as follows:

The first term, the indirect effect, is the price change for uncontrolled locations due to the end of rent control, which is in turn due to maintenance and allocative effects as in equation ðA2Þ. The second term, the direct effect, is the price change in a controlled economy going from a rent-controlled location to an uncontrolled location. For a formerly controlled location, the direct effect of the end of rent control is larger when the controlled price at the location is further depressed. The following proposition summarizes the relevant considerations from this model.

Proposition 1. When rent control ends, the price change for uncontrolled locations is greater for neighborhoods

• with a larger fraction of locations with rent control ðl↑Þ,

• where the price of controlled locations is further depressed from their market price ðk 1 ↓Þ, and

• where there is greater misallocation of resident types relative to the types in the uncontrolled economy ðk 2 ↓Þ.

Furthermore, when rent control ends, controlled locations experience an additional price increase due to the direct effect of decontrol.

This model shows the difficulty involved in distinguishing between direct and indirect effects at decontrolled locations. When rent control ends, there is a direct price effect due to the formerly controlled location being priced by the market. However, there is also an indirect effect as neighborhood amenities improve as a result of increases in maintenance and the income of residents, leading to higher prices. This in turn leads a landlord to invest in additional maintenance. For empirical purposes, at decontrolled locations, the direct and indirect channels cannot be readily distinguished because each affects the equilibrium level of the other.

The model's simplicity also imposes some limitations for our setting. First, the price of housing services is an abstraction that allows for no distinction between house prices and rents, which might be especially relevant in a dynamic setting. The model does not therefore allow for realistic dynamics to capture expectations of neighborhood appreciation and the option value of ownership. Second, amenities within a neighborhood are assumed to be pure public goods, so residents have no desire to substitute between locations within a neighborhood. If housing services were instead differentiated, there might be substitution between different locations within a neighborhood. In this case, new construction stimulated by the end of rent control might have a price impact at nearby uncontrolled housing ðdue to increased housing supplyÞ. Third, residents are identical in the model except for their income: within a neighborhood, all residents at uncontrolled locations ðthough not generally at controlled locationsÞ have the same level of income and, because of spatial arbitrage, obtain the same utility. Finally, the model focuses on one housing market and does not consider neighboring markets that do not have rent control. Although it is not modeled, it is possible that residents at previously controlled locations move out of Cambridge and that residents in these neighboring towns move into Cambridge with the end of rent control. 1991-94 1995-2004 1991-94 1995-2004 1991-94 1995-2004 1991-94 1995-2004 A. Permits Note.-Data source is the universe of Cambridge Inspectional Services permits issued during 1991-2004. If a structure receives multiple permits in a given year, we sum these expenditures and treat them as a single permit. When calculating units permitted or expenditures per unit in a year, we attribute the structure's permitted status and expenditures to only one unit. Expenditures are converted to real 2008 dollars using the Consumer Price Index for All Items Less Shelter for All Urban Consumers, Series Id: CUUR0000SA0L2, Not Seasonally Adjusted.

annex

Note.-Samples includes houses ðpanel AÞ and condominiums ðpanel BÞ that transacted in 1994 and 2004. Regression models follow col. 5 of table 4. In cols. 4-6, the dependent variable is the assessed value of any unit that is on a map-lot at which at least one unit transacted in the given year. The number of observations for houses is larger in cols. 1-3 than in cols. 4 -6 because a unit may transact more than once per year. The number of observations for condominiums in cols. 4 -6 is larger than in cols. 1-3 because condominium structures contain multiple units. We observe the market price for transacted units and the assessed price for all units in the structure but cannot determine which specific unit in a structure has transacted. In specifications that include RC, RCI, non-RC Â RCI or RC Â RCI interacted with Post, main effects of these variables are also included but not tabulated. Robust standard errors clustered by 1990 block group are in parentheses. See the note to table 6

Funding

invaluable access to expertise and data.We acknowledge generous support from the Alfred P. Sloan Foundation, the Lincoln Institute for Land Policy, the National Science Foundation ðgrant SES-962572Þ, and the Rappaport Institute for Greater Boston.Palmer thanks the National Science Foundation Graduate Research Fellowship ðgrant 0645960Þ.We received excellent research assistance from

Failed to load PDF:

StripeM-Inner

Paper sources

Abstract screening pilot

Abstract screening results

Extraction pilot

Extraction results

Research report

Modify setup