The Federal Reserve Board eagle logo links to home page
Skip to: [Printable Version (PDF)] [Bibliography] [Footnotes]
Finance and Economics Discussion Series: 2008-51 Screen Reader version ♣

Property Taxes and Elderly Labor Supply

Hui Shan *
October 23, 2008
Keywords: Property tax, property tax relief programs, elderly, labor supply

Abstract:
The recent housing market boom in the U.S. has caused sharp increases in residential property taxes. Anecdotal evidence suggests that rising property taxes have induced elderly homeowners to increase their labor supply. This paper uses 1992-2004 panel data from the Health and Retirement Study (HRS) as well as a newly collected dataset on state-provided property tax relief programs to investigate the effect of property taxes on the labor supply of elderly homeowners. It is the first rigorous study on the link between property taxes and elderly labor supply. I examine both the extensive margin--whether elderly homeowners delay retirement or reenter the labor market in the face of rising property taxes, and the intensive margin--whether elderly homeowners work longer hours when property taxes increase. A simulated IV approach is used to address the potential endogeneity problem associated with property taxes. I find little evidence that property taxes have a significant impact on elderly homeowners' decisions to retire, to re-enter the labor force, or to increase working hours.

JEL classification: H31, H71, J14, J22

1 Introduction

During the late 1990s and early 2000s, the U.S. housing market experienced a remarkable boom, which led to sharp increases in residential property taxes. U.S. Census data indicate that from 2000 to 2005, median housing values went up by 50% and median property taxes rose by 30% in real terms. Anecdotal evidence suggests that such unexpected rises in property taxes may induce elderly homeowners, especially those housing-rich but income-poor elderly homeowners, to increase their labor supply by delaying retirement. Unfortunately, there have been no systematic studies investigating the link between property taxes and elderly labor supply. This paper serves as the first attempt to study this link.

Property taxes may potentially influence elderly labor supply through two channels: wealth effects and liquidity constraints. Economists have long recognized that unexpected changes in wealth may induce individuals to adjust their labor supply. Because increases in property taxes are equivalent to declines in wealth, they may lead elderly homeowners to consume less leisure and supply more labor. Property taxes may also cause elderly homeowners to increase their labor supply because of liquidity constraints. For example, according to the 1992-2004 Health and Retirement Study (HRS) sample, 10% of the homeowners between age 50 and 75 reported paying 9% or more of their income for property taxes. For 25% of these homeowners, annual property tax payments represented at least 40% of household financial assets. The lack of liquid assets among many elderly homeowners make them vulnerable to increases in property taxes. Therefore, they may resort to delaying retirement, reentering the labor force, and/or working longer hours in order to stay in their homes.

On the other hand, property taxes may not have a significant impact on elderly labor supply for two reasons. First, the reduction in wealth caused by rising property taxes may be small relative to elderly homeowners' total wealth. For example, consider a 60 year old who expects to live in his house for another 20 years. Even if annual property taxes increase permanently by $500, he only experiences a $6,731 wealth decline in present discounted value (PDV) assuming a 5% discount rate. Previous studies such as [1] and [2] find that on average, a $100,000 increase in wealth causes retirement rates to decline by 10%. A reduction of several thousand dollars in wealth simply may not be enough to trigger noticeable changes in retirement and labor force reentry behavior. Second, elderly homeowners may respond to rising property taxes by relocating to low-tax areas or by downsizing to smaller houses rather than by increasing labor supply. In fact, Shan (2008) finds evidence suggesting that property taxes raise the mobility rate among elderly homeowners. If elderly homeowners have already lowered their property tax burdens by moving to low-tax areas or smaller houses, it may no longer be necessary for them to increase their labor supply at the same time.

In this paper, I empirically test the relationship between property taxes and elderly labor supply. Specifically, I use panel data from the 1992-2004 Health and Retirement Study (HRS) and newly-collected data on state-provided property tax relief programs to estimate the property tax effect on the labor supply of homeowners aged 50-75. In particular, I focus on three labor supply outcome variables in my regression analysis: retirement, reentry to the labor market, and working hours. Because property tax payments may be endogenous to individuals' labor supply decisions, I exploit the variation in state-provided property tax relief programs and construct simulated relief benefits as instruments for property taxes. Such simulated relief benefits measure the generosity of property tax relief programs and thus, are negatively correlated with property tax payments. The simulation procedure makes sure that these instruments contain only the variation in program rules and depend exclusively on state, age, and year. To the extent that state, age, and year are exogenous, these simulated instruments satisfy the exclusion restriction. The central IV estimation results cannot reject that property taxes have no significant impact on elderly homeowners' decisions to retire, to reenter the labor force, or to increase working hours. Such findings imply that incidences reported in news articles where elderly homeowners have been delaying retirement to keep up with rising property taxes are unlikely to be representative. Elderly homeowners may have chosen to move rather than to increase labor supply in their effort to reduce property tax burdens.

This paper contributes to the property tax literature and the wealth effect literature in several ways. First, to my knowledge, it is the first study to look at how property taxes affect labor supply. Property taxes are the most important tax revenue source for local governments, and property tax relief programs cost about $10 billion annually in the United States.1 Studying the behavioral impact of property taxes on elderly homeowners is indispensable for any normative analysis of property taxes and property tax relief programs. Second, previous research studying the wealth effect on retirement behavior has exploited variations in Social Security and pension benefits, stock market booms and busts, housing market movements, inheritances, and lottery winnings. This paper complements the existing literature by using property taxes and property tax relief programs as a novel source of variation. Third, while most existing studies focus only on retirement behavior, this paper examines both the extensive margin - whether rising property taxes induce elderly homeowners to delay retirement or reenter the labor force, and the intensive margin - whether elderly homeowners work longer hours when property taxes increase. By looking beyond retirement decisions, this paper provides more comprehensive evidence on how wealth shocks affect elderly labor supply.

The rest of this paper proceeds as follows. Section 2 reviews the existing literature on the wealth effect and introduces the background on property taxes. In section 3, I describe the HRS data used in this paper. I explain my empirical strategy, discuss the estimation results, and present robustness checks and extensions in section 4. The last section concludes and points out some caveats of this paper.


2 Background

A sizable literature exists on the wealth effect and retirement behavior. As mentioned above, previous studies have relied on variations in Social Security and pension benefits, stock market movements, housing market movements, inheritances, and lotteries for identification. Earlier works on retirement incentives of Social Security benefits, including Diamond and Hausman (1984), Burtless (1986), Krueger and Pischke (1992), and Blau (1994), generally find that even though the effect of Social Security is statistically significant, it is small relative to the trend toward early labor force exit among older men. More recent works adopt the "option value" approach developed by Stock and Wise (1990) and estimate the dynamic effect of Social Security and pensions on retirement decisions. Samwick (1998), Chan and Stevens (2004), and Coile and Gruber (2007) implement such dynamic models and show that forward-looking incentive measures for Social Security and private pensions are significant determinants of retirement.

The stock market boom and bust as well as the remarkable housing value run-up in recent years have provided researchers arguably exogenous sources of variation for studying the wealth effect on retirement behavior. Using the HRS data, Coronado and Perozek (2003) find that individuals who held corporate equity immediately before the bull market of the 1990s on average retired earlier than those who did not. Sevak (2005) compares individuals with defined contribution pension plans and individuals with defined benefit pension plans. She finds that unexpected gains in wealth during the 1990s bull market induced earlier retirement. Using the HRS, Current Population Survey (CPS), and Survey of Consumer Finances (SCF), Coile and Levine (2006) exploited both the stock market boom in the late 1990s and the stock market bust in the early 2000s to study the impact of wealth shocks on retirement decisions. They find that the stock market has very little influence on aggregate labor market behavior. Farnham and Sevak (2007) and Goodstein (2008) use cross-MSA variation in housing price movements to identify the wealth effect on retirement timing. They find that increases in housing wealth raise the probability of retirement significantly.

In search for exogenous sources of variation to measure the wealth effect, researchers have also estimated the effect of inheritance receipt and lottery winning on labor supply. Brown, Coile and Weisbenner (2006) show that inheritance receipt is associated with a significant increase in the probability of retirement, and the effect is stronger when the inheritance is unexpected. Imbens, Rubin and Sacerdote (2001) use an original survey of people playing the lottery in Massachusetts in the mid-1980s and find that wealth shocks reduce labor supply.

In summary, the existing literature generally supports the theoretical prediction that labor supply responds to wealth shocks. Nevertheless, the evidence shown in the literature is far from conclusive. Studies that employ difference-in-differences frameworks rely heavily on the assumption that in absence of the wealth effect, treatment groups and control groups would have the same propensity to retire conditional on covariates. Such an assumption may be too strong in many cases. Even for studies that have reasonably tight identification strategies, the magnitudes of estimated wealth effects vary considerably from one study to another. As tens of millions of baby-boomers approach retirement age in coming years, the field calls for more research to provide new evidence on this important subject. This paper uses variations in property taxes and property tax relief programs to estimate the wealth effect on elderly labor supply.

Property taxes are responsible for approximately 72% of all local tax revenues, representing the most important tax revenue source for local governments. In 2004, property tax collections in the U.S. exceeded $300 billion.2 The housing market boom of the late 1990s and early 2000s led to significant increases in residential property taxes. Such steep rises in property taxes may be more burdensome to elderly homeowners than to non-elderly homeowners for two reasons. First, the current U.S. tax system allows taxpayers who choose to itemize their deductions on federal income tax returns to deduct property tax payments. Because mortgage interest payments are usually the main reason for choosing itemized deductions over standard deductions, and because elderly homeowners are likely to have paid off their mortgages and take standard deductions, the marginal cost of paying an extra dollar in property tax is usually higher for elderly homeowners than for non-elderly homeowners. As a result, elderly homeowners may have to increase their labor supply in order to stay in their homes.

Second, many elderly homeowners do not have substantial liquid assets to cover rising property taxes. Table 1 displays the present discounted values (PDV) of hypothetical property tax increases for homeowners for different ages and discount rates. For simplicity, I assume away longevity risks and impose that everyone lives to age 80 exactly. The PDV as a percentage of median household financial wealth is shown in parenthesis. The top panel illustrates the case where annual property taxes experience a permanent increase of $300. For example, if the homeowner paid $2,000 for property taxes last year, he would be paying $2,300 for property taxes every year from now on. Even though the PDV of such an increase in property taxes is only a few thousand dollars, it represents 16-47% of the median household financial wealth. The bottom panel of Table 1 illustrates the case where property taxes increase by $300 annually. Using the previous example, the homeowner who paid $2,000 for property taxes last year would pay $2,300 this year, $2,600 next year, and so on. The PDV of such an increase in property taxes overwhelms the median household financial wealth.

However, there are also reasons to be skeptical that increasing property taxes have generated a large impact on elderly homeowners' labor supply. First, the wealth effect alone may not induce noticeable changes in labor supply behavior. The top panel of Table 1 shows that even for a homeowner of age 50 with a low discount rate of 0.02, a hypothetical property tax increase of $300 only amounts to a $7,019 reduction in wealth. Given that previous studies on wealth effects suggest that a $100,000 increase in wealth raises retirement rates by roughly 10%, a one-time permanent increase in property taxes is unlikely to generate much wealth effect on elderly homeowners. Elderly homeowners may respond to rising property taxes by increasing labor supply only if they expect property taxes continue to rise in coming years as illustrated in the bottom panel of Table 1.

The second reason why the relationship between property taxes and elderly labor supply may not be empirically detectable is that elderly homeowners have an alternative strategy to reduce their property tax payments, namely, by moving to low-tax area or by downsizing. As noted above, Shan (2008) finds evidence suggesting that rising property taxes induce higher mobility among elderly homeowners. If the disutility from delaying retirement, reentering labor force, or working longer hours outweighs the transaction cost associated with moving and downsizing, we may not find a significant impact of property taxes on elderly labor supply. For these reasons, the theoretical prediction of the degree to which property taxes may affect elderly labor supply is ambiguous, and we have to rely on empirical studies to determine the relationship between property taxes and elderly labor supply.


3 Data Description

The data used in this paper has two components: the Health and Retirement Study (HRS) and newly collected data on property tax relief programs. The HRS is a biannual panel of a nationally representative sample of elderly and near-elderly individuals in the United States. At present, seven waves of the survey (1992-2004) have been released to researchers. The HRS includes households from four different cohorts: the HRS cohort (born between 1931 and 1941), the AHEAD cohort (born before 1924), the "Children of the Depression" (CODA) cohort (born between 1924 and 1930), and the "War Baby" (WB) cohort (born between 1942 and 1947).3 The HRS cohort appears in all seven waves. The AHEAD cohort was first interviewed in 1993 and then in 1995. Since 1998, the AHEAD cohort has been interviewed concurrently with the HRS cohort biannually. The CODA and WB cohorts appear only in the last four waves (1998-2004). The raw dataset has 26,867 individuals and 126,104 person-wave observations.

The HRS data have detailed information on demographics, health, labor supply, and finances. Whenever possible, I use the RAND HRS Data File, a user-friendly version that contains a subset of HRS variables.4 Because this paper examines the wealth effect on elderly labor supply, I limit the sample to individuals of age between 50 and 75. Figures 1 and 2 display the relationship between age and labor force status for male and female HRS respondents, respectively. At age 50, around 85% of males were in the labor force and only 10% of males were out of the labor force.5 In contrast, almost 30% of females were out of the labor force at age 50. As people grew older, the fraction of respondents remaining in the labor force declined. Females appeared to exit the labor force earlier than males. For both males and females, the biggest jump in retirement occurred at age 62. This is probably because 62 is the early retirement age at which beneficiaries can claim Social Security benefits. At age 75, only 5% of females remained in the labor force, whereas over 10% of males were still in the labor force. Because of these apparent differences in male and female labor supply behavior, I perform regression analysis for older men and older women separately. Figure 3 plots the empirical retirement hazard rate for homeowners between age 50 and 75. Conditional on being in the labor force, the probability that one retires within the next two years goes up with age. For both males and females, the hazard rate increases sharply around age 60 and again around age 70.

Figure 1:
Figure 1: Fraction of Males in Labor Force by Age: One curve shows the fraction of male individuals who are working or unemployed by age. This curve starts at around 0.1 at age 50 and increases to around 0.9 at age 75. The second curve shows the fraction of male individuals who are retired or out of labor force by age. This curve starts at around 0.9 at age 50 and decreases to around 0.1 at age 75. The two curves are mirror images of each other and cross at age 62. A vertical line marks age 62.


Figure 2:
Figure 2: Fraction of Females in Labor Force by Age: One curve shows the fraction of female individuals who are working or unemployed by age. This curve starts at around 0.3 at age 50 and increases to around 0.95 at age 75. The second curve shows that fraction of female individuals who are retired or out of labor force by age. This curve starts at around 0.7 at age 50 and decreases to around 0.05 at age 75. The two curves are mirror images of each other and cross at age 61. A vertical line marks age 62.


Figure 3:
Figure 3: Retirement Hazard Rate of Homeowners by Age: The figure plots the retirement hazard rates for male and female homeowners respectively by age. The two curves follow each other closely. They increase between age 50 and 61. They then stay relatively flat between age 61 and 68. Finally, they increase again between age 68 and 75.



To measure changes in labor supply, I use three outcome variables: retirement, reentry to the labor force, and working hours. I define retirement as a transition from working or being unemployed to being retired or out of the labor force. Similarly, a transition from being retired or out of the labor force to working or being unemployed is defined as reentry to the labor force. Working hours refer to the self-reported total number of hours worked during the past year. Table 2 shows that on average, the two-year retirement rate is 18.3% for males and 20.4% for females in the sample. The average two-year reentry rate is much lower: 5.8% for males and 5.2% for females. Conditional on being in the labor force, male respondents report an average of 2,283 annual working hours, and female respondents report an average of 1,880 annual working hours.

The key independent variable in this paper is property taxes. In all seven waves, respondents were asked to report the amount of property taxes paid on their primary residence during the past year. I assume these self-reported property tax payments are the actual payments after all relevant property tax exemptions, rebates or refunds have been applied. Such an assumption is crucial for the first-stage regression in my IV strategy. For programs where participation is automatic and property tax bills are mailed to homeowners after benefits have been netted out, this assumption seems justified. For programs where homeowners receive rebate checks soon after paying property taxes, it is unclear whether respondents report their before-relief property tax payments or after-relief property tax payments. For programs that are implemented by state personal income tax credits, respondents are likely to report their before-relief benefits for two reasons. First, relief benefits are usually received long after homeowners have paid their property taxes. Second, property tax relief benefits may appear less salient on state personal income tax returns. For example, filers may view property tax credits that they claim against income tax liabilities as income tax relief benefits rather than property tax relief benefits. Recent studies including Chetty, Looney and Kroft (2007) and Finkelstein (2007) suggest that tax salience could have a significant impact on behavior. Therefore, I exclude in my regression analysis states where relief benefits are granted by tax credits on state personal income tax returns.6 The dropped observations represent about 25% of the sample. I also drop individuals living in mobile homes and individuals living on farms or ranches because these properties may be treated differently from other residential properties for tax purposes.

Table 2 displays summary statistics of key demographic and socio-economic variables for the retirement, reentry, and working-hour samples. Unsurprisingly, individuals in the retirement and working-hour samples are younger, healthier, better-educated, and have significantly higher household income than individuals in the reentry sample. Individuals in the retirement and working-hour samples also live in more expensive houses and pay higher property taxes than individuals in the reentry sample. On the other hand, they have lower financial wealth than their counterparts in the reentry sample. Such a pattern in housing wealth and financial wealth may suggest that homeowners transform their housing wealth into financial wealth by downsizing as they age and exit the labor force.

In addition to the publicly available HRS data, I obtained restricted access to household-level geographic identifiers in each survey year, including state, county, census tract, and zip code. The state identifier is crucial in my analysis because it links households with the state-provided property tax relief programs for which they are eligible. The county identifier allows me to control for county-year specific unemployment rate published by the Census Bureau in my regression analysis.

The second component of the data used in this paper is the data on property tax relief programs. As of the present, all 50 states and the District of Columbia have some form of property tax relief programs for homeowners, especially for low-income and elderly homeowners. Many of these programs were first established well before my sample period started. Broadly speaking, there are four categories of relief programs: Homestead Exemptions and Credits, Circuit-Breakers, Deferral Programs, and Limitations. Shan (2008) has detailed descriptions on how these programs work, how the data were collected, and how these programs are codified. At the end of the process, a computer program is written to produce three output variables: the amount of benefits from homestead exemption, homestead credit, and circuit-breaker programs that a homeowner is eligible for, whether eligible for an "assessment value freeze" program, and whether eligible for an "property tax freeze" program. Such output variables can be generated for any homeowner in the U.S. in any year between 1990 and 2004 provided that input parameters, including state of residence, year, age, income, house value, Social Security income, marital status, household size and wealth, are non-missing.


4 Empirical Strategy and Estimation Results

In this section, I present the empirical model and estimation results in studying the effect of property taxes on elderly homeowners' decisions to retire, to reenter the labor force, and to increase working hours. Estimations are performed for men and women separately. Robustness checks and extensions are carried out and discussed at the end of this section.


4.1 Property Taxes and Retirement Decisions

To investigate whether property taxes have an impact on retirement behavior, I start with a simple probit model7

Prob\displaystyle (Retire_{ist}=1) = \Phi (\beta_1 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t) (1)

where Retire_{ist} indicates whether household i in state s retired between time t and t+1, \zeta_s denotes state fixed effects, \delta_t denotes year fixed effects, and the covariate vector \mathbf{X}_{ist} includes a constant, income quintile indicators, house value quintile indicators, financial wealth quintile indicators, race/ethnicity (i.e. White, black, and Hispanic), whether married, education categories (i.e. less than high school, high school graduates, some college, and college graduates), whether hospitalized between the last interview and the current interview, whether have pension coverage, whether have retiree health insurance coverage, county unemployment rate, industry dummies, occupation dummies, and age dummies.8 The key variable of interest in equation (1) is Tax_{ist}, property tax payments by household i in state s at time t. If higher property taxes cause elderly homeowners to delay retirement, then we expect \beta_1 < 0.

Columns (1) and (3) in Table 3 present estimation results of equation (1) for males and females, respectively. To make the results interpretable, I show marginal effects of independent variables by calculating the predicted marginal effect for each observation and then averaging them across all observations. To be consistent with results presented later in this section, standard errors shown in parentheses are bootstrapped by 500 random draws with replacement. I implement a block-bootstrap scheme to make certain that observations are clustered at state level in estimating standard errors. The estimated effects of property taxes are negative as expected, but statistically insignificant. The magnitudes of the marginal effects are small, suggesting that a $100 increase in annual property taxes is associated with a 0.03 percentage point decrease in two-year retirement rate for men and 0.09 percentage point decrease for women.

There are three reasons why such estimates of \beta_1 may be inconsistent. First, property taxes are products of tax rates and house values. At a given tax rate, higher house values lead to higher property taxes. If house values affect elderly homeowners' labor supply decisions through channels other than property taxes (e.g. housing wealth effect), then the probit estimate of \beta_1 will be biased to the extent that house values are not fully controlled for. Second, property taxes are used to provide local public services. Higher property taxes often correlate with better local public services. If local public services such as parks and senior centers are complements to the consumption of leisure, we will not be able to estimate \beta_1 consistently without controlling for local public services which are unobservable to econometricians. Lastly, property tax payments are self-reported in the HRS. To the extent that elderly homeowners do not know and/or report property taxes accurately, measurement errors will cause attenuation bias in estimating \beta_1. To deal with these three problems, I use measures of property tax relief program generosity to instrument for property taxes.

More specifically, I use the set of instruments - \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} - that are described in detail in Shan (2008). Because property tax relief benefits reduce property tax payments, these measures of program generosity should be negatively correlated with property tax payments. Such a negative correlation serves as the first stage in this paper. On the other hand, these instruments essentially capture variations in property tax relief program rules and are rid of variations stemming from individual characteristics. Thus, they are orthogonal to the individual level error term \epsilon_{ist} and satisfy the exclusion restriction. Table 2 illustrates the summary statistics of \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist}. In the retirement sample, 5.0% of males and 9.3% of females are eligible for relief benefits from homestead exemptions, homestead credits, and circuit-breakers. Conditional on eligibility, the average benefits from these programs are $144 for males and $202 for females. In addition, 7.5% of both males and females are eligible for assessment value freeze programs. 13.4% of males and 14.7% of females are eligible for property tax freeze programs.

To implement the simulated IV strategy in a probit framework, I use the two-step estimator suggested by Rivers and Vuong (1988).9 Beside computational ease, the Rivers-Vuong two-step IV approach has another appealing feature. The usual probit t-test on \hat{v}, which is a consistent estimate of the first-stage error term, is a valid test of the null hypothesis that Tax_{ist} is exogenous. Such a test is equivalent to the Hausman specification test suggested by Hausman (1978). Because I use a two-step procedure to estimate the IV-probit model, standard errors need to be adjusted accordingly. I choose to obtain consistent estimates of standard errors by bootstrapping in lieu of the delta-method for two reasons. First, bootstrapping is computationally easier to implement. Second, bootstrapping provides higher-order refinements while the delta-method is only a first-order approximation (Horowitz (2001)).

Columns (2) and (4) of Table 3 show the IV-probit estimation results using \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} as instruments. The estimated marginal effects of property taxes remain negative and statistically insignificant. The magnitudes of these marginal effects become much larger than the probit results. They suggest that a $100 increase in annual property tax payments reduces the two-year retirement rate by 0.71 percentage points for men and 1.35 percentage points for women. Given the average two-year retirement rate of 18.3% for men and 20.4% for women, these represent a 3.9 percent decline in retirement rate for men and 6.6 percent decline for women. Although the point estimates imply a sizable property tax effect on retirement behavior, the standard errors are large and we cannot reject the null hypothesis that property taxes do not affect retirement. Note that the first-stage F-statistic is only 2.10 for the male sample and 14.28 for the female sample. The reason why the instruments do not strongly correlate with property tax payments for males is probably that male respondents in the retirement sample tend to have significantly higher household income than females. In addition, males tend to be older than their spouses. Since the age requirement in property tax relief programs often refers to the oldest person in the household, male homeowners are less likely to qualify for relief benefits than female homeowners of the same age. High incomes and not having older spouses may have prevented male homeowners from taking advantages of property tax relief programs which often have income and age as qualification criteria. Stock, Wright and Yogo (2002) suggest that the rule of thumb for detecting weak instruments is to check whether the first-stage F-stat exceeds 10. By this standard, the male sample may have a weak instrument problem and the IV-probit estimates may be biased in the direction of the probit estimates. Moreover, the Hausman test rejects the null hypothesis that property tax payments are exogenous in the female sample but not in the male sample.

The estimated marginal effects of the other covariates are mostly consistent with our expectation and previous literature's findings. For example, health shocks, approximated by the indicator variable "whether the respondent was recently hospitalized," raise the two-year retirement rate by 5 percentage points for both men and women, or a 25 percent increase from the baseline level. Financial wealth is correlated with higher probability of retirement. However, such a correlation should not be interpreted as causal since individuals who have strong desires to retire early may have saved more aggressively over their life-cycle. In addition, male respondents who have retiree health insurance coverage are more likely to retire than those who do not, but the effect is insignificant for females. Female respondents who have pension coverage are less likely to retire than those who do not, but the effect is insignificant for males. Black and Hispanic women are more likely to retire than white women, although race/ethnicity does not appear to matter among male respondents. Such differences between males and females highlight the importance of analyzing male and female individuals separately in studying labor supply behavior.


4.2 Property Taxes and Reentry Decisions

In the previous section, I estimate a retirement regression model and the results cannot reject the null hypothesis that property taxes do not have a significant effect on elderly homeowners' retirement decisions. In this section, I explore the impact of property taxes on labor force reentry behavior in a similar regression analysis by estimating the following probit model:

Prob\displaystyle (Reentry_{ist}=1) = \Phi (\beta_2 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t) (2)

where Reentry_{ist} indicates whether individual i who is out of the labor force at time t reenters the labor force between time t and t+1. If higher property taxes cause retired elderly homeowners to reenter the labor force, then we expect \beta_2 > 0.

I again use \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} as instruments for property taxes to obtain consistent estimates of \beta_2. As shown in Table 2, individuals in the reentry sample are relatively older and have lower household income because they have to be out the labor force to be in this sample. As a result, they are more likely to be eligible for property tax relief programs that target low-income and elderly homeowners. On average, 22.5% of males and 27.2% of females in the reentry sample are eligible for homestead exemptions, homestead credits, or circuit-breakers. 10.1% of males and 10.8% of females are eligible for assessment value freeze programs, and 17.6% of males and 18.1% of females are eligible for property tax freeze programs. The average two-year reentry rate among homeowners age 50-75 is low, 5.8% for males and 5.2% for females.

Table 4 presents estimation results of both probit and IV-probit specifications for males and females separately. For the male sample, the estimated marginal effect of property taxes is positive but statistically insignificant in the probit specification. The marginal effect doubles in the IV-probit specification, but remains statistically indistinguishable from zero. For the female sample, both the probit and IV-probit specifications produce negative estimates of \beta_2, and the marginal effects of property taxes on reentry behavior are also statistically insignificant. The first-stage relationship between property taxes and the instruments are strong, with a F-statistic of 138.47 for the male sample and 11.09 for the female sample. All told, the evidence does not support the claim that homeowners who face higher property taxes are more likely to reenter the labor force.

Estimation results displayed in Table 4 also suggest that both male and female Hispanic homeowners are more likely to reenter the labor force than white and black elderly homeowners. When county unemployment rate is high, older men and women are less likely to reenter the labor force. Higher income is correlated with higher probability of reentry behavior, especially among male homeowners. Among female homeowners, individuals who live in more expensive houses are more likely to reenter the labor force. Among male homeowners, individuals with more financial wealth are less likely to reenter the labor force. Moreover, negative health shocks appear to prevent older men from reentering the labor force. Married women are less likely to reenter the labor force than their unmarried counterparts.


4.3 Property Taxes and Working Hours

The previous two sections have examined the property tax effect on the extensive margin of elderly labor supply, namely, whether to exit or reenter the labor market. In this section, I investigate the intensive margin of labor supply by estimating the effect of property taxes on whether elderly homeowners' working hours. I employ a regression model in the following form:

\displaystyle Hours_{ist} = \beta_3 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t + \epsilon_{ist} (3)

where Hours_{ist} is the total number of hours individual i reports working at time t conditional on being in the labor force. If higher property taxes indeed induce elderly homeowners to work longer hours, we expect \beta_3 >0.

As before, because Tax_{ist} may be endogenous to individuals' labor supply decisions and cause bias in estimating \beta_3, I use measures of property tax relief program generosity, \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist}, to instrument for Tax_{ist}. Note that the first stage relationship between property tax payments and property tax relief program generosity may be weak in the working-hour regression. Elderly homeowners have to be in the labor force at both time t and t+1 to be considered in this analysis. Hence, individuals in the working-hour sample are relatively young and have higher household income. Such characteristics imply that they tend to be ineligible for property tax relief programs that are designed to help low-income and older homeowners. Therefore, the correlation between property tax payments and property tax relief program generosity may disappear. For example, table 2 shows that on average only 4.7% of males and 9.0% of females in the working-hour sample are eligible for homestead exemptions, homestead credits, and circuit-breakers. 7.3% of male and 7.3% of females are eligible for assessment value freeze programs, and 13.3% of males and 14.6% of females are eligible for property tax freeze programs. On average, the male respondents report to work 2,283 hours annually and the female respondents report to work 1,880 hours annually.

Table 5 presents the estimation results of the OLS and 2SLS specifications for males and females separately. In the male samples, the OLS estimate suggests that property taxes have a positive, small, and statistically insignificant effect on working hours. In the female sample, however, the OLS estimate suggests that property taxes have a negative and statistically significant effect on working hours. Such a counterintuitive result may reflect that property taxes are endogenous to labor supply decisions. For instance, homeowners who have strong preferences for local amenities such as parks and senior centers also prefer consuming more leisure and work fewer hours. If they choose to live in areas with high property taxes and better local amenities, we would observe a negative correlation between property tax payments and working hours. Once property taxes are instrumented using relief program generosity measures, the effect of property taxes on working hours appears to be negative and statistically insignificant for both the male and female sample. The estimated coefficients are large, but the standard errors are also large and I cannot reject the null hypothesis that property taxes have no impact on elderly homeowners' working hours. Similar to the retirement analysis, the first-stage relationship between property taxes and the instruments is weak for males in the working-hour sample, probably because they have high incomes and they tend to have younger spouses. On the other hand, the first stage F-statistic is 18.31 for the females, suggesting that I do not have a weak-instrument problem in the female sample. Nevertheless, the 2SLS estimate of the coefficient on property taxes does not support the hypothesis that higher property taxes induce elderly homeowners to work longer hours.

Results shown in Table 5 also suggest that income is highly correlated with working hours. In the male sample, black homeowners work fewer hours than white and Hispanic homeowners. In the female sample, homeowners with higher financial wealth appear to work fewer hours. Married women work fewer hours than women with other marital status. Women with college degrees work more hours than women with less education. Female homeowners living in counties with high unemployment rates work slightly fewer hours compared with those in counties with low unemployment rates.


4.4 Robustness Checks and Extensions

In previous sections, I have used a simulated IV strategy to identify the potential effect of property taxes on elderly homeowners' labor supply decisions both on the extensive margin and the intensive margin. The estimation results suggest that property taxes may have no significant impact on elderly homeowners' decision to retire, to reenter the labor force, or to increase working hours. In this section, I first carry out robustness checks by using various sub-samples. Then I extend the regression models and allow property taxes to differentially affect the labor supply decisions of homeowners at different ages. Because the weak-instrument problem may exist in the male retirement sample and the male working-hour sample, I focus on females when analyzing retirement and working-hour responses, and I look at both males and females when studying reentry behavior.

As shown in Figure 1 and Figure 2, most elderly homeowners exit the labor market between age 55 and 70. In the first robustness check, I limit the sample to homeowners of age 55-70 and investigate whether the estimates change once homeowners younger than 55 or older than 70 are dropped. In the second robustness check, I exclude elderly homeowners who live in California because Proposition 13 may have created a very unusual institutional setting. Proposition 13 was adopted in California in 1978. It limits property tax rates at 1% and requires assessment values to grow no more than 2% per year unless the house is sold and re-assessment is carried out. In the third robustness check, I drop individuals who claim to be self-employed because self-employed individuals may face higher or lower costs than others when adjusting their labor supply. Lastly, I exclude elderly homeowners who report having moved between time t and t+1 and focus on individuals who stay in the same house in both periods.

Table 6, 7 and 8 present the estimation results using these sub-samples in the retirement, reentry, and working-hour regressions, respectively. In the retirement analysis, the estimated marginal effect of property taxes is negative across sub-samples for female respondents, which is consistent with the hypothesis that rising property taxes induce elderly homeowners to delay retirement. However, none of the estimates is statistically different from zero at conventional confidence levels, and thus, I cannot reject the null hypothesis that property taxes have no significant impact on retirement behavior. In the reentry analysis, the estimated coefficient on property taxes is positive in some cases and negative in others. In addition, they are all statistically indistinguishable from zero. Therefore, there appears to be little evidence that elderly homeowners who are out of the labor force actually reenter the labor force in order to boost their incomes and pay for rising property taxes. In the working-hour analysis, most estimates of the property tax effect are negative, which is inconsistent with the notion that higher property taxes may have caused elderly homeowners to work longer hours. Additionally, none of the estimates are statistically significant.

Next, I allow for heterogeneity in the property tax effect for individuals of different ages. If the reason why I do not find evidence of significant property tax effect on elderly labor supply is that I have restricted the coefficients on property taxes to be the same for individuals of different ages, then this extension should be able to identify the age groups at which property taxes may have a noticeable impact on labor supply decisions. Because Figure 3 shows that retirement hazard rate increases sharply at age 60, and because age 62 and 65 are the Social Security early retirement age and normal retirement age respectively, I allow the coefficients on property taxes to differ across five age groups: 50-59, 60, 61-62, 63-65, and 66-75. Specifically, I estimate the following probit models:

Prob\displaystyle (Retire_{ist}=1) \displaystyle = \displaystyle \Phi (\alpha_1 Tax_{ist}\cdot \mathbf{1}(Age<60) + \alpha_2 Tax_{ist}\cdot \mathbf{1}(Age=60) (4)
  \displaystyle + \displaystyle \alpha_3 Tax_{ist}\cdot \mathbf{1}(Age=61-62) + \alpha_4 Tax_{ist}\cdot \mathbf{1}(Age=63-65)  
  \displaystyle + \displaystyle \alpha_5 Tax_{ist}\cdot \mathbf{1}(Age>65) + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t)  


Prob\displaystyle (Reentry_{ist}=1) \displaystyle = \displaystyle \Phi (\gamma_1 Tax_{ist}\cdot \mathbf{1}(Age<60) + \gamma_2 Tax_{ist}\cdot \mathbf{1}(Age=60) (5)
  \displaystyle + \displaystyle \gamma_3 Tax_{ist}\cdot \mathbf{1}(Age=61-62) + \gamma_4 Tax_{ist}\cdot \mathbf{1}(Age=63-65)  
  \displaystyle + \displaystyle \gamma_5 Tax_{ist}\cdot \mathbf{1}(Age>65) + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t)  


\displaystyle Hours_{ist} \displaystyle = \displaystyle \lambda_1 Tax_{ist}\cdot \mathbf{1}(Age<60) + \lambda_2 Tax_{ist}\cdot \mathbf{1}(Age=60) (6)
  \displaystyle + \displaystyle \lambda_3 Tax_{ist}\cdot \mathbf{1}(Age=61-62) + \lambda_4 Tax_{ist}\cdot \mathbf{1}(Age=63-65)  
  \displaystyle + \displaystyle \lambda_5 Tax_{ist}\cdot \mathbf{1}(Age>65) + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t + \epsilon_{ist}  

where \mathbf{1}(\cdot) returns one if the expression in parenthesis holds true and zero otherwise. Since Tax_{ist} is endogenous to individual i's labor supply decisions, I use the interactions between the five age group dummies and the three program generosity measures, \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist}, to instrument for the interactions between the five age group dummies and Tax_{ist}. Thus, I have 5 endogenous explanatory variables and 15 instruments in each equation.

Table 9 shows the estimation results of IV-probit and 2SLS specifications. Many of the estimated coefficients on property taxes have signs inconsistent with the hypothesis that rising property taxes induce elderly homeowners to increase their labor supply. Among the ones that have the expected signs, none of the IV estimates is statistically different from zero at conventional significance level. Overall, the above extension, where heterogenous property tax effects are allowed, does not detect a systematic relationship between property taxes and elderly labor supply.

In summary, despite efforts to identify the link between property taxes and elderly labor supply using various sub-samples and allowing for heterogeneous effects across age groups, there appears to be little evidence suggesting that property taxes play a significant role in elderly homeowners' labor supply decisions. Note that the instruments used in this paper to identify the causal effect of property taxes - simulated relief benefits from homestead exemptions, homestead credits, and circuit-breakers, eligibility for assessment value freeze programs, and eligibility for property tax freeze programs - affect property taxes of only homeowners who are eligible for property tax relief programs and actually take up these programs. To the extent that these people are more sensitive and responsive to property taxes, the estimates presented here may provide the upper bound of the property tax impact on elderly labor supply. Therefore, finding little evidence supporting the claim that elderly homeowners respond to rising property taxes by increasing labor supply in this paper implies that property taxes probably play an insignificant role in labor supply decisions of the general public.


5 Conclusion

Property taxes are the most important tax revenue source of local governments in the United States. The recent housing market boom led to substantial increases in property taxes which in turn have caught the attention of both policy makers and the general public. News articles have reported anecdotes of elderly homeowners delaying retirement in the face of rising property taxes, but, until now, there has been no empirical study on the relationship between property taxes and elderly labor supply. Exploiting the arguably exogenous variation in state-provided property tax relief programs, this paper is the first study that examines the role property taxes play in elderly homeowners' labor supply decisions. I examine both the extensive and intensive margins of labor supply behavior. This paper also contributes to the existing literature on the wealth effect by using property taxes and property tax relief programs as a novel source of variation. Various sub-samples are analyzed and the property tax effect is allowed to differ across age groups. Overall, I find little evidence supporting the claim that elderly homeowners have been delaying retirement, reentering the labor force, or working longer hours to deal with increasing property taxes.

There are two caveats worth mentioning. First, I have to focus on people who are in the labor force in order to study retirement and working-hour behavior. This limits the power of my instruments significantly in the retirement and working-hour regressions among male respondents because these people are often too young and their incomes tend to be too high for them to be eligible for property tax relief programs. Since weak instruments may bias the IV estimates, the lack of a significant estimated relationship between retirement and working-hour responses and property taxes in the male sample in the IV-probit specification does not completely rule out the possibility that property taxes play an important role in older men's retirement and working-hour decisions. On the other hand, the first-stage is quite strong in the reentry analysis. Nevertheless, there appears to be little evidence of labor force reentry response to property taxes.

Second, even though this paper find evidence suggesting that the wealth shock generated by unexpected increases in property taxes may not have a significant effect on elderly labor supply, it does not necessarily mean that the wealth effects on elderly labor supply are in general insignificant. Perhaps increases in property taxes do not translate into a large enough reduction in wealth and therefore, do not induce a detectable effect on labor supply. Furthermore, even if homeowners truly do not respond to the negative wealth shock produced by rising property taxes, they may still be very responsive to other forms of wealth shocks such as stock market and housing market booms and busts. More research is needed on whether and to what degree different types of wealth affect elderly labor supply differently.

Taken together with Shan (2008), the findings of this paper have important policy implications. Shan (2008) shows evidence suggesting that higher property taxes induce higher mobility rates among elderly homeowners. Property taxes may affect elderly mobility through various channels: the wealth effect, the liquidity constraint effect, and the substitution effect. The wealth effect exists because increases in property taxes are equivalent to declines in total wealth. The liquidity constraint effect means that elderly homeowners would have preferred staying in their homes if they were able to afford rising property taxes. The only reason that they move in response to higher property taxes is that they have no incomes or liquid assets to pay for increases in property taxes. The substitution effect refers to the fact that elderly homeowners, who typically do not have school-age children living in the house, often find the marginal cost of paying high property taxes exceeds the marginal benefit of consuming local public services such as schools. Thus, increases in property taxes may trigger an adjustment in their choice of housing consumption bundles, and such an adjustment is usually accomplished by moving. These different mechanisms have different welfare implications. Although Shan (2008) shows the relationship between property taxes and elderly mobility, she does not identify whether this relationship is driven by the wealth effect, the liquidity constraint effect, or the substitution effect. On the other hand, property taxes affect elderly labor supply only through the wealth effect and the liquidity constraint effect. Finding little evidence supporting that property taxes play a significant role in elderly homeowners' labor supply decisions, this paper points in the direction of property taxes influencing elderly mobility through the substitution effect. In this case, property tax relief programs may have kept elderly homeowners in their homes when they optimally should have moved to areas with lower property taxes and fewer public services.


Bibliography

Blau, David. "Labor Force Dynamics of Older Men."
Econometrica 62(1), 1994, 117-156.
Brown, Jeffrey, Courtney Coile, and Scott Weisbenner. "The Effect of Inheritance Receipt on Retirement."
NBER Working Paper #12386, 2006.
Bradley, David. "Property Taxes in Perspective."
Center on Budget and Policy Priorities, March, 2005.
Burtless, Gary. "Social Security, Unanticipated Benefit Increases, and the Timing of Retirement."
Review of Economic Studies 53(176), 1986, 781-805.
Chan, Sewin and Ann Huff Stevens. "Do Changes in Pension Incentives Affect Retirement? A Longitudinal Study of Subjective Retirement Expectations."
Journal of Public Economics 88(7-8), 2004, 1307-1333.
Chetty, Raj, Adam Looney and Kory Kroft. "Salience and Taxation: Theory and Evidence."
Working Paper, 2007.
Coile, Courtney and Jonathan Gruber. "Future Social Security Entitlements and the Retirement Decision."
Review of Economics and Statistics 89(2), 2007, 234-246.
Coile, Courtney and Phillip Levine. "Bulls, Bears, and Retirement Behavior."
Industrial and Labor Relations Review 59(3), 2006, 408-429.
Coronado, Julia and Maria Perozek. "Wealth Effects and the Consumption of Leisure: Retirement Decision During the Stock Market Boom of the, 1990s."
Federal Reserve Board Finance and Economics Discussion Series, 2003-20.
Diamond, Peter and Jerry Hausman. "Retirement and Unemployment Behavior of Older Men."
Retirement and Economic Behavior edited by Henry Aaron and Gary Burtless. Washington D.C.: Brookings Institution, 1984.
Farnham, Martin and Purvi Sevak. "Housing Wealth and Retirement Timing."
Michigan Retirement Research Center Working Paper, 2007-172.
Finkelstein, Amy. "E-Z Tax: Tax Salience and Tax Rates."
Working Paper, 2007.
Goodstein, Ryan. "The Effect of Wealth on Labor Force Participation of Older Men."
University of North Carolina - Chapel Hill Dissertation Chapter, 2008.
Hausman, Jerry. "Specification Tests in Econometrics."
Econometrica 46(6), 1978, 1273-1291.
Horowitz, Joel. "The Bootstrap."
Handbook of Econometrics, Vol 5., edited by J.J. Heckman and E. Leamer, Elsevier Science B.V., 2001, Ch 52, 3159-3228.
Imbens, Guido and Joshua Angrist. "Identification and Estimation of Local Average Treatment Effects."
Econometrica 62(2), 1994, 467-475.
Krueger, Alan and Jorn-Steffan Pischke. "The Effect of Social Security on Labor Supply: A Cohort Analysis of the Nortch Generation."
Journal of Labor Economics 10(4), 1992, 412-437.
Lyons, Karen, Sarah Farkas, and Nicholas Johnson. "The Property Tax Circuit Breaker: An Introduction and Survey of Current Programs."
Center on Budget and Policy Priorities, March, 2007.
"A Guide to Property Taxes: Series."
National Conference of State Legislatures, 2005.
Rivers, Douglas and Quang Vuong. "Limited Information Estimators and Exogeneity Tests for Simultaneous Probit Models."
Journal of Econometrics 39(3), 1988, 347-366.
Samwick, Andrew. "New Evidence on Pensions, Social Security, and the Timing of Retirement."
Journal of Public Economics 70(2), 1998, 207-236.
Sevak, Purvi. "Wealth Shocks and Retirement Timing: Evidence from the Nineties."
Michigan Retirement Research Center Working Paper, 2002-027, Revised 2005.
Shan, Hui. "Property Taxes and Elderly Mobility."
Massachusetts Institute of Technology Dissertation Chapter, 2008.
St. Clair, Patricia, et al. "RAND HRS Data Documentation, Version F."
Labor & Population Program, RAND Center for the Study of Aging, May, 2006.
Stock, James and David Wise. "Pensions, the Option Value of Work, and Retirement."
Econometrica 58(5), 1990, 1151-1180.
Stock, James, Jonathan Wright and Motohiro Yogo. "A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments."
Journal of Business and Economics Statistics 20(4), 2002, 518-529.



Table 1, Part 1: PDV of Hypothetical Property Tax Increases:
(A One-Time Permanent Increase of $300)
Discount Rate Age=50 Age=60 Age=70
0.02 7,019 4,912 3,677
0.02 PDV (47%) (33%) (25%)
0.05 5,205 4,039 3,245
0.05 PDV (35%) (27%) (22%)
0.08 2,995 2,617 2,313
0.08 PDV (20%) (18%) (16%)

Table 1, Part 2: PDV of Hypothetical Property Tax Increases:
Increases of $300 Each Year
Discount Rate Age=50 Age=60 Age=70
0.02 101,253 60,110 38,091
0.02 PDV (685%) (406%) (258%)
0.05 53,491 37,324 26,918
0.05 PDV (362%) (252%) (182%)
0.08 17,376 14,429 12,119
0.08 PDV (117%) (98%) (82%)

Notes: I assume the individual lives to age 80. Numbers in parenthesis represent the PDV as a percentage of median household financial wealth among homeowners of that age. Based on the 1992-2004 HRS data, the median household financial wealth is $14,790 for homeowners of age 50, $27,472 for homeowners of age 60, and $29,171 for homeowners of age 70.



Table 2: Summary Statistics of Analysis Samples
Retirement Sample Male Mean Male Median Male SD Female Mean Female Median Female SD
Retire 0.183   0.387 0.204   0.403
Simulated Benefits: Fraction Eligible 0.050   0.148 0.093   0.163
Simulated Benefits: Conditional Benefits 144 99.5 190 202 154.4 172
Value Freeze 0.075   0.263 0.075   0.264
Tax Freeze 0.134   0.340 0.147   0.355
Property Tax 1,839 1,307 2,149 1,621 1,220 1,727
Household Income 103,582 72,436 179,602 76,915 59,159 104,146
House Value 175,393 132,055 222,679 155,066 121,564 122,653
Financial Wealth 125,771 23,320 689,661 85,662 19,144 223,768
Age 57.5 57 4.5 57.2 57 4.4
Black 0.057   0.231 0.068   0.252
Hispanic 0.055   0.228 0.051   0.219
Married 0.876   0.330 0.697   0.459
Recently Hospitalized 0.132   0.339 0.119   0.323
Less than High School 0.162   0.369 0.136   0.343
High School Graduates 0.279   0.448 0.337   0.473
Some College 0.215   0.411 0.286   0.452
College Graduates 0.344   0.475 0.241   0.428
Pension Coverage 0.665   0.472 0.601   0.490
Retiree Health Insurance 0.460   0.498 0.334   0.472

Table 2: Summary Statistics (Continued)

Reentry Sample Male Mean Male Median Male SD Female Mean Female Median Female SD
Reentry 0.058   0.235 0.052   0.221
Simulated Benefits            
Fraction Eligible 0.225   0.312 0.272   0.304
Conditional Benefits 179 138 160 194 147 157
Value Freeze 0.101   0.302 0.108   0.310
Tax Freeze 0.176   0.381 0.181   0.385
Property Tax 1,455 1,056 1,520 1,436 1,000 3,381
Household Income 58,901 40,171 75,595 57,313 35,243 123,611
House Value 154,180 119,650 132,410 150,789 110,000 190,515
Financial Wealth 164,095 38,134 415,897 155,324 34,334 433,659
Age 66.0 66 6.2 64.6 65 6.8
Black 0.062   0.241 0.059   0.236
Hispanic 0.044   0.204 0.058   0.234
Married 0.842   0.364 0.729   0.444
Recently Hospitalized 0.247   0.431 0.188   0.391
Less than High School 0.255   0.436 0.223   0.417
High School Graduates 0.275   0.447 0.382   0.486
Some College 0.201   0.401 0.229   0.420
College Graduates 0.269   0.443 0.166   0.372

Table 2: Summary Statistics (Continued)

Working-Hour Sample Male Mean Male Median Male SD Female Mean Female Median Female SD
Working Hours 2,283 2,100 639 1,880 2,080 665
Simulated Benefits            
Fraction Eligible 0.047   0.141 0.090   0.160
Conditional Benefits 142 98 189 200 154 170
Value Freeze 0.073   0.261 0.073   0.261
Tax Freeze 0.133   0.339 0.146   0.353
Property Tax 1,910 1,317 4,149 1,619 1,207 1,688
Household Income 101,912 71,912 168,523 77,651 58,210 113,154
House Value 173,267 132,055 169,279 155,439 121,491 125,985
Financial Wealth 121,845 23,048 653,379 92,258 19,144 421,381
Age 57.5 57 4.4 57.2 57 4.4
Black 0.058   0.233 0.067   0.251
Hispanic 0.060   0.237 0.052   0.223
Married 0.876   0.329 0.702   0.457
Recently Hospitalized 0.137   0.344 0.120   0.325
Less than High School 0.167   0.373 0.141   0.348
High School Graduates 0.277   0.447 0.343   0.475
Some College 0.219   0.413 0.285   0.451
College Graduates 0.338   0.473 0.231   0.422

Note: One has to be working or unemployed at time t to be included in the retirement sample and the working hours sample. One has to be retired or out of labor force at time t to be included in the reentry sample. Property tax, household income, house value, and financial wealth are in 2000 dollars. Individual weights are applied.



Table 3: Retirement Estimation Result
  Male (1) Profit Male (2) IV-Profit Female (3) Provit Female (4) IV-Profit
Property Taxes (in 10,000) -0.0264 -0.7057 -0.0905 -1.3460
Property Taxes (in 10,000) (Standard Error) (0.0342) (0.9739) (0.0580) (1.1622)
Income Quintile 2 -0.0407* -0.0319 -0.0505** -0.0499**
Income Quintile 2 (Standard Error) (0.0230) (0.0274) (0.0237) (0.0236)
Income Quintile 3 -0.0375* -0.0332 -0.0166 -0.0080
Income Quintile 3 (Standard Error) (0.0215) (0.0231) (0.0249) (0.0263)
Income Quintile 4 -0.0191 -0.0206 -0.0031 0.0037
Income Quintile 4 (Standard Error) (0.0230) (0.0230) (0.0275) (0.0277)
Income Quintile 5 -0.0444* -0.0205 -0.0081 0.0288
Income Quintile 5 (Standard Error) (0.0237) (0.0401) (0.0301) (0.0478)
House Value Quintile 2 0.0141 0.0285 0.0054 0.0414
House Value Quintile 2 (Standard Error) (0.0165) (0.0233) (0.0211) (0.0385)
House Value Quintile 3 -0.0174 0.0151 -0.0020 0.0738
House Value Quintile 3 (Standard Error) (0.0183) (0.485) (0.0217) (0.0714)
House Value Quintile 4 -0.0053 0.0535 -0.0031 0.1123
House Value Quintile 4 (Standard Error) (0.0204) (0.0841) (0.0244) (0.1082)
House Value Quintile 5 -0.0253 0.1270 -0.0179 0.2221
House Value Quintile 5 (Standard Error) (0.0230) (0.2216) (0.0277) (0.2222)
Financial Wealth Quintile 2 0.0198 0.0157 0.0418** 0.0296
Financial Wealth Quintile 2 (Standard Error) (0.0157) (0.0177) (0.0175) (0.0211)
Financial Wealth Quintile 3 0.0289* 0.0289 0.0748*** 0.0652***
Financial Wealth Quintile 3 (Standard Error) (0.0166) (0.0179) (0.0202) (0.0228)
Financial Wealth Quintile 4 0.0535*** 0.0510** 0.0914*** 0.0805***
Financial Wealth Quintile 4 (Standard Error) (0.0174) (0.0201) (0.0199) (0.0239)
Financial Wealth Quintile 5 0.0930*** 0.1268** 0.1249*** 0.1494***
Financial Wealth Quintile 5 (Standard Error) (0.0194) (0.0548) (0.0229) (0.0417)
Black 0.0169 0.0112 0.0619*** 0.0562**
Black (Standard Error) (0.0199) (0.0220) (0.0202) (0.0225)
Hispanic -0.0562** -0.0505 0.0378 0.0471*
Hispanic (Standard Error) (0.0248) (0.0313) (0.0270) (0.0273)
Married -0.0309* -0.0315 0.0194 0.0057
Married (Standard Error) (0.0184) (0.0267) (0.0161) (0.0231)
High School Graduate -0.0220 -0.0200 -0.0240 -0.0151
High School Graduate (Standard Error) (0.0157) (0.0181) (0.0163) (0.0181)
Some College -0.0177 -0.0165 -0.0300 -0.0146
Some College (Standard Error) (0.0179) (0.0199) (0.0187) (0.0235)
College Graduate -0.0237 -0.0038 -0.0307 0.0051
College Graduate (Standard Error) (0.0203) (0.0392) (0.0225) (0.0389)
Recently Hospitalized 0.0479*** 0.0507*** 0.0527*** 0.0522**
Recently Hospitalized (Standard Error) (0.0138) (0.0165) (0.0170) (0.0203)
Pension Coverage -0.0147 -0.0036 -0.0606*** -0.0721***
Pension Coverage (Standard Error) (0.0118) (0.0191) (0.0140) (0.0240)
Retiree Health Insurance 0.0487*** 0.0502*** 0.0085 0.0036
Retiree Health Insurance (Standard Error) (0.0103) (0.0140) (0.0139) (0.0149)
County Unemployment Rate 0.0015 0.0013 0.0000 -0.0025
County Unemployment Rate (Standard Error) (0.0025) (0.0026) (0.0029) (0.0039)
First Stage F-stat   2.10   14.28
Hausman Test   3.0650   5.4105***
(coeff on first-stage residuals)   (5.8850)   (1.5272)
N 6,388 6,388 5,657 5,657
Pseudo R2 0.1489   0.1093  

Note: The regression model is Prob (Retire_{ist}=1) = \Phi (\beta_1 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t). Other than the variables shown in the table, \mathbf{X}_{ist} also includes a constant, age dummies, industry dummies, and occupation dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist} in the IV-probit specifications. The numbers shown in the table are marginal effects averaged across observations. Standard errors in parentheses are bootstrapped by 500 random draws with replacement clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significa nt at 0.01 level.



Table 4: Reentry Estimation Result
  Male (1) Probit Male (2) IV-Probit Female (3) Probit Female (4) IV-Probit
Property Taxes (in 10,000) 0.0263 0.0581 -0.0030 -0.7037
Property Taxes (in 10,000) (Standard Error) (0.0188) (0.4371) (0.0155) (0.4619)
Income Quintile 2 0.0253** 0.0255** 0.0156* 0.0156
Income Quintile 2 (Standard Error) (0.0103) (0.0112) (0.0081) (0.0134)
Income Quintile 3 0.0422*** 0.0427*** 0.0238*** 0.0240
Income Quintile 3 (Standard Error) (0.0113) (0.0125) (0.0086) (0.0166)
Income Quintile 4 0.0334*** 0.0338** 0.0222** 0.0213
Income Quintile 4 (Standard Error) (0.0120) (0.0134) (0.0090) (0.0173)
Income Quintile 5 0.0870*** 0.0859*** 0.0284*** 0.0519**
Income Quintile 5 (Standard Error) (0.0130) (0.0250) (0.0102) (0.0224)
House Value Quintile 2 0.0032 0.0026 0.0077 0.0288*
House Value Quintile 2 (Standard Error) (0.0109) (0.0151) (0.0089) (0.0152)
House Value Quintile 3 0.0025 0.0007 0.0151 0.0578**
House Value Quintile 3 (Standard Error) (0.0115) (0.0254) (0.0093) (0.0248)
House Value Quintile 4 -0.0201 -0.0231 0.0073 0.0775*
House Value Quintile 4 (Standard Error) (0.0125) (0.0393) (0.0102) (0.0469)
House Value Quintile 5 0.0021 -0.0045 0.0110 0.1512*
House Value Quintile 5 (Standard Error) (0.0135) (0.0747) (0.0111) (0.0860)
Financial Wealth Quintile 2 0.0048 0.0049 0.0082 0.0111
Financial Wealth Quintile 2 (Standard Error) (0.0106) (0.0115) (0.0086) (0.0113)
Financial Wealth Quintile 3 -0.0213* -0.0216* 0.0068 0.0149
Financial Wealth Quintile 3 (Standard Error) (0.0112) (0.0120) (0.0088) (0.0123)
Financial Wealth Quintile 4 -0.0393*** -0.0395*** -0.0017 0.0063
Financial Wealth Quintile 4 (Standard Error) (0.0119) (0.0129) 0.0093) (0.0166)
Financial Wealth Quintile 5 -0.0465*** -0.0479*** -0.0243** -0.0126
Financial Wealth Quintile 5 (Standard Error) (0.0127) (0.0172) (0.0101) (0.0183)
Black -0.0107 -0.0103 0.0111 0.0115
Black (Standard Error) (0.0127) (0.0145) (0.0091) (0.0125)
Hispanic 0.0323** 0.0322** 0.0188* 0.0338**
Hispanic (Standard Error) (0.0131) (0.0147) (0.0109) (0.0153)
Married 0.0106 0.0104 -0.0227*** -0.0305***
Married (Standard Error) (0.0097) (0.0115) (0.0071) (0.0103)
High School Graduate -0.0102 -0.0104 -0.0093 -0.0106
High School Graduate (Standard Error) (0.0092) (0.0098) (0.0070) (0.0097)
Some College 0.0022 0.0020 0.0015 0.0079
Some College (Standard Error) (0.0097) (0.0115) (0.0074) (0.0116)
College Graduate -0.0122 -0.0133 0.0004 0.0222
College Graduate (Standard Error) (0.0104) (0.0195) (0.0092) (0.0206)
Recently Hospitalized -0.0285*** -0.0283*** -0.0023 0.0015
Recently Hospitalized (Standard Error) (0.0087) (0.0097) (0.0062) (0.0107)
County Unemployment Rate -0.0031* -0.0032* -0.0023** -0.0039**
County Unemployment Rate (Standard Error) (0.0017) (0.0019) (0.0011) (0.0017)
First Stage F-stat   138.47   11.09
Hausman Test   -0.314   5.285
(coeff on first-stage residuals)   (2.430)   (3.630)
N 6,475 6,475 9,406 9,406
Pseudo R2 0.1538   0.1391  

Note: The regression model is Prob (Reentry_{ist}=1) = \Phi (\beta_2 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t). Other than the variables shown in the table, \mathbf{X}_{ist} also includes a constant and age dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist} in the IV-probit specifications. The numbers shown in the table are marginal effects averaged across observations. Standard errors in parentheses are bootstrapped by 500 random draws with replacement clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level



Table 5: Working Hour Estimation Result
  Male (1) OLS Male (2) 2SLS Female (3) OLS Female (4) 2SLS
Property Taxes (in 10,000) 36.4 -776.7 -164.9** -3069.7
Property Taxes (in 10,000) (Standard Error) (23.9) (2044.0) (80.8) (2967.2)
Income Quintile 2 116.0*** 98.2 142.7*** 146.3**
Income Quintile 2 (Standard Error) (24.9) (62.6) (44.2) (53.7)
Income Quintile 3 168.9*** 139.6* 178.3*** 195.6***
Income Quintile 3 (Standard Error) (29.6) (81.2) (46.0) (63.0)
Income Quintile 4 198.6*** 161.3 230.3*** 241.2***
Income Quintile 4 (Standard Error) (28.7) (109.9) (51.2) (63.9)
Income Quintile 5 262.7*** 262.3** 271.6*** 348.6***
Income Quintile 5 (Standard Error) (34.9) (65.1) (51.9) (112.4)
House Value Quintile 2 -20.6 -2.5 -47.6 29.1
House Value Quintile 2 (Standard Error) (31.7) (69.8) (42.5) (100.6)
House Value Quintile 3 -43.5* 16.1 -9.7 155.5
House Value Quintile 3 (Standard Error) (21.9) (160.0) (42.3) (197.5)
House Value Quintile 4 -23.6 74.7 -45.8 212.9
House Value Quintile 4 (Standard Error) (31.1) (270.3) (37.4) (289.5)
House Value Quintile 5 30.0 227.8 -78.3* 462.0
House Value Quintile 5 (Standard Error) (43.9) (529.2) (40.5) (557.7)
Financial Wealth Quintile 2 - 35.2 25.9 -8.4 -28.3
Financial Wealth Quintile 2 (Standard Error) (22.9) (37.5) (31.4) (35.8)
Financial Wealth Quintile 3 - 36.1 43.1 -25.6 -28.8
Financial Wealth Quintile 3 (Standard Error) (28.5) (32.5) (30.7) (36.0)
Financial Wealth Quintile 4 26.5 32.0 -78.1** -84.3**
Financial Wealth Quintile 4 (Standard Error) (34.0) (37.3) (31.7) (39.0)
Financial Wealth Quintile 5 -0.5 45.4 -123.0** -32.3
Financial Wealth Quintile 5 (Standard Error) (34.8) (110.0) (45.2) (107.1)
Black -86.8** -94.7** -30.6 -10.7
Black (Standard Error) (34.5) (44.2) (28.9) (44.0)
Hispanic -58.9 -62.9 -35.8 6.3
Hispanic (Standard Error) (44.5) (50.2) (41.5) (65.3)
Married -22.8 -17.5 -201.7*** -219.1***
Married (Standard Error) (32.4) (34.9) (18.4) (27.8)
High School Graduate -17.0 -2.7 -15.0 3.9
High School Graduate (Standard Error) (24.3) (33.2) (37.1) (37.0)
Some College -6.4 -2.0 -13.0 17.8
Some College (Standard Error) (31.1) (27.1) (47.0) (55.6)
College Graduate -35.2 -6.8 114.5** 194.4**
College Graduate (Standard Error) (37.6) (64.7) (52.5) (87.0)
Recently Hospitalized -36.1 -46.6 5.2 4.1
Recently Hospitalized (Standard Error) (25.2) (43.6) (29.9) (38.5)
County Unemployment Rate 4.6 6.6 -8.9** -15.3*
County Unemployment Rate (Standard Error) (4.2) (9.3) (3.8) (7.6)
First Stage F-stat   0.90   18.31
Hausman Test   813   2906
(coeff on first-stage residuals)   (1851)   (2713)
N 7,442 7,442 6,552 6,552
Pseudo R2 0.3289 . 0.3009 .

Notes: The regression model is Hours_{ist} = \beta_3 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t + \epsilon_{ist} . Other than the variables shown in the table, \mathbf{X}_{ist} also includes a constant and age dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist} in the IV-probit specifications. The numbers shown in the table are marginal effects averaged across observations. Standard errors in parentheses are bootstrapped by 500 random draws with replacement clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level.



Table 6: Robustness Checks of the Retirement Regression Analysis - Female Sample
  Original Sample (1) Age 55-70 (2) Drop CA (3) Drop Self-Emp (4) Drop Movers (5)
Property Tax (in 10,000s) -1.346 -2.148* -1.302 -1.161 -1.525
Property Tax (in 10,000s) (Standard Error) (1.162) (1.225) (1.025) (0.812) (1.090)
First Stage F-stat 14.3 52.4 31.0 53.2 13.2
Hausman Test 5.411*** 9.141*** 4.938*** 4.477*** 6.644***
Hausman Test (Standard Error) (1.527) (1.852) (1.429) (1.654) (2.495)
N 5,657 4,173 5,016 4,895 5,159

Notes: The regression model is Prob (Retire_{ist}=1) = \Phi (\beta_1 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t). \mathbf{X}_{ist} includes a constant, income quintile indicators, house value quintile indicators, financial wealth quintile indicators, race/ethnicity dummies, whether married, education categories, whether recently hospitalized, whether have pension coverage, whether have retiree health insurance coverage, county unemployment rate, industry dummies, occupation dummies, and age dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist}. The numbers shown in the table are marginal effects averaged across observations. Standard errors in parentheses are bootstrapped by 500 random draws with replacement clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level.



Table 7: Robustness Checks of the Reentry Regression Analysis - Male Sample
Male Sample Original Sample (1) Age 55-70 (2) Drop CA (3) Drop Self-Emp (4) Drop Movers (5)
Property Tax (in 10,000s) 0.058 -0.588 0.183 -0.043 0.067
Property Tax (in 10,000s) (Standard Error) (0.437) (0.795) (0.423) (0.441) (0.551)
First Stage F-stat 138.5 59.4 157.4 137.7 74.3
Hausman Test -0.314 4.183 -1.339 0.674 -0.448
Hausman Test (Standard Error) (2.430) (4.977) (2.278) (2.877) (3.133)
N 6,475 4,396 5,684 5,879 5,848

Table 7: Robustness Checks of the Reentry Regression Analysis - Female Sample
Female Sample Original Sample (1) Age 55-70 (2) Drop CA (3) Drop Self-Emp (4) Drop Movers (5)
Property Tax (in 10,000s) -0.704 -0.283 -0.609 -0.501 -0.497
Property Tax (in 10,000s) (Standard Error) (0.462) (0.399) (0.404) (0.439 ) (0.457)
First Stage F-stat 11.1 21.9 24.9 10.2 9.1
Hausman Test 5.285 2.318 4.474 4.137 3.847
Hausman Test (Standard Error) (3.630) (4.642) (3.177) (4.422) (4.197)
N 9,406 6,485 8,282 9,095 8,550

Notes: Note: The regression model is Prob (Reentry_{ist}=1) = \Phi (\beta_2 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t). \mathbf{X}_{ist} includes a constant, income quintile indicators, house value quintile indicators, financial wealth quintile indicators, race/ethnicity dummies, whether married, education categories, whether recently hospitalized, county unemployment rate, and age dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist}. The numbers shown in the table are marginal effects averaged across observations. Standard errors in parentheses are bootstrapped by 500 random draws with replacement clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level.



Table 8: Robustness Checks of the Working Hour Regression Analysis - Female Sample
  Original Sample (1) Age 55-70 (2) Drop CA (3) Drop Self-Emp (4) Drop Movers (5)
Property Tax (in 10,000s) -3070 -3378 -3233 1076 -593
Property Tax (in 10,000s) (Standard Error) (2967) (2406) (3045) (927) (3456)
First Stage F-stat 18.3 49.7 18.6 75.2 11.7
Hausman Test 2906 3193 3076 -1168 442
Hausman Test (Standard Error) (2713) (2208) (2696) (949) (3487)
N 6,552 4,810 5,812 5,647 5,997

Notes: The regression model is Hours_{ist} = \beta_3 Tax_{ist} + \mathbf{X}_{ist}\mathbf{\Pi} + \zeta_s + \delta_t + \epsilon_{ist} . \mathbf{X}_{ist} includes a constant, income quintile indicators, house value quintile indicators, financial wealth quintile indicators, race/ethnicity dummies, whether married, education categories, whether recently hospitalized, county unemployment rate, industry dummies, occupation dummies, and age dummies. \zeta_s is state fixed effects. \delta_t is year fixed effects. \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for Tax_{ist}. Standard errors in parentheses are clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level.



Table 9: Property Tax Effect on Homeowners of Different Age Groups
  Retirement Female (1) IV-Profit Reentry Male (2)
IV-Profit		 Reentry Female (3) IV-Profit Hours Female (4) 2SLS
PropTax*(Age <60) 0.042 0.119 0.983 -219
PropTax*(Age <60) (Standard Error) (0 .984) (0.650) (0.759) (2093)
PropTax*(Age=60) 2.158 -0.251 -1.247 -2989
PropTax*(Age=60) (Standard Error) (1.530) (1.314) (4.063) (3275)
PropTax*(Age=61-62) -0.628 0.443 -2.819** -3231
PropTax*(Age=61-62) (Standard Error) ( 1.057) (1.448) (1.367) (3567)
PropTax*(Age=63-65) 1.275 0.630 0.101 1648
PropTax*(Age=63-65)(Standard Error) (1.791) (1.657) (1.345) (2679)
PropTax*(Age >65) -0.585 0.188 -0.445 255
PropTax*(Age >65) (Standard Error) (0.749) (0.779) (0.997) (1055)
N 5,657 6,475 9,406 6,552

Notes: Other controls include a constant, income quintile indicators, house value quintile indicators, financial wealth quintile indicators, race/ethnicity dummies, whether married, education categories, whether recently hospitalized, whether have pension coverage, whether have retiree health insurance coverage, county unemployment rate, industry dummies, occupation dummies, and age dummies, state fixed effects, and year fixed effects. The interactions between the five age groups and \widetilde{Benefits}_{ist}, ValueFreeze_{ist}, and TaxFreeze_{ist} are used as instruments for the interactions between the five age groups and property taxes. Standard errors in parentheses are clustered at state level. Individual weights from HRS are applied. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level.



Footnotes

* Division of Research and Statistics, Board of Governors of the Federal Reserve System, Washington, D.C. 20551, [email protected]. I am very grateful to Jim Poterba, Bill Wheaton and Jerry Hausman for advice and support. I thank Tonja Bowen Bishop, Amanda Kowalski and David Powell for helpful discussions. I also thank Mohan Ramanujan for technical assistance with using HRS restricted data. This research was supported by the National Institute on Aging, Grant Number P01-AG05842, and the Lincoln Institute of Land Policy. The findings and conclusions expressed are solely those of the authors and do not represent views of the Board of Governors, the staff of the Federal Reserve System, or the National Institute of Aging, or the Lincoln Institute. Return to Text
1. Author's estimate using 2004 data reported in Lyons, Farkas and Johnson (2007). Return to Text
2. See Bradley (2005) and NCSL (2005). Return to Text
3. In 2004, a fifth cohort, Early Boomers (born between 1948 and 1953), was added to the HRS. Because households in this cohort have only been interviewed once and I need at least two adjacent surveys to study whether this period's property taxes affect labor supply between this period and the next period, I exclude them from my analysis. Return to Text
4. See St.Clair et al (2006) for more information on the RAND HRS Data File. Return to Text
5. The other 5% of male respondents were disabled or had missing labor force status. Return to Text
6. These states are District of Columbia, Massachusetts, Michigan, Missouri, Montana, New Jersey, New Mexico, New York, Oklahoma, Rhode Island, Vermont, and Wisconsin. I do not exclude states that use rebate checks to implement relief programs because the sample size would drop significantly and asymptotic theory no longer applies when there are only a few states left in the sample and standard errors are clustered at the state level. Return to Text
7. I use a probit model in this paper because the mean of dependent variables is not near 0.5. A linear probability model may be biased when the dependent variable is close to zero or one, and will produce predictions beyond the range of zero to one. Return to Text
8. For the first wave in 1992, HRS asked whether the individual was hospitalized in the past year. From the second wave on, HRS asked whether the individual was hospitalized since the last interview. Return to Text
9. The Rivers-Vuong two-step approach is a limited information procedure. Thus, it is less efficient than the conditional maximum likelihood estimation (MLE). In practice, I find MLE computationally difficult, and iterations do not converge. Return to Text


♣ This version is optimized for use by screen readers. Descriptions for all mathematical expressions are provided in LaTex format. A printable pdf version is available. Return to Text
Home | Economic Research and Data | Accessibility | Contact us
Last update: October 29, 2008