Podcast
Questions and Answers
Assuming a constant marginal propensity to consume and utilizing a permanent income hypothesis framework, how would the observed stability of the wealth effect on earnings, as depicted in Figure 1, be affected if individuals systematically underestimated the persistence of their lottery winnings?
Assuming a constant marginal propensity to consume and utilizing a permanent income hypothesis framework, how would the observed stability of the wealth effect on earnings, as depicted in Figure 1, be affected if individuals systematically underestimated the persistence of their lottery winnings?
- The observed effect would exhibit increased volatility, characterized by unpredictable spikes and drops reflecting belief revisions.
- The observed effect would be amplified initially, followed by a gradual reversion to zero as individuals update their beliefs.
- The observed effect would remain unchanged, as the permanent income hypothesis posits rational expectations and perfect foresight.
- The observed effect would be dampened initially, with a subsequent increase as individuals realize the permanence of their winnings. (correct)
Within the context of the provided study, if a significant portion of lottery winners experienced a simultaneous, exogenous shock that negatively impacted their non-labor income specifically, how would this influence the interpretation of Figure 1 regarding the wealth effect on labor earnings, assuming labor and leisure are normal goods?
Within the context of the provided study, if a significant portion of lottery winners experienced a simultaneous, exogenous shock that negatively impacted their non-labor income specifically, how would this influence the interpretation of Figure 1 regarding the wealth effect on labor earnings, assuming labor and leisure are normal goods?
- It would not affect the interpretation, as the difference-in-differences design controls for any common shocks affecting lottery winners.
- It would lead to an overestimation of the true wealth effect, as individuals increase labor supply to compensate for the loss in non-labor income. (correct)
- It would lead to an underestimation of the true wealth effect, as individuals decrease labor supply due to the combined effect of the lottery winnings and the negative shock.
- It would create a spurious correlation, making it impossible to isolate the true wealth effect from the effect of the non-labor income shock.
Assuming lottery winnings are subject to a progressive tax system with bracket creep, and individuals fail to fully anticipate this effect when making labor supply decisions, how would the observed labor supply response to lottery wealth be affected in the years following the win?
Assuming lottery winnings are subject to a progressive tax system with bracket creep, and individuals fail to fully anticipate this effect when making labor supply decisions, how would the observed labor supply response to lottery wealth be affected in the years following the win?
- The labor supply response would become more elastic, as individuals become more sensitive to changes in their marginal tax rate.
- The reduction in labor supply would be smaller than anticipated, as the higher tax bracket partially offsets the wealth effect. (correct)
- The reduction in labor supply would be larger than anticipated, as individuals attempt to maintain their pre-lottery disposable income level.
- There would be no change in labor supply, as the progressive tax system only affects individuals with very high incomes.
Counterfactually, imagine the Swedish government introduced a universal basic income (UBI) scheme shortly after the lottery winnings were distributed. How would this contemporaneous policy intervention likely affect the estimated wealth effect on labor earnings, and what econometric strategy could best isolate the lottery wealth effect from the UBI's impact?
Counterfactually, imagine the Swedish government introduced a universal basic income (UBI) scheme shortly after the lottery winnings were distributed. How would this contemporaneous policy intervention likely affect the estimated wealth effect on labor earnings, and what econometric strategy could best isolate the lottery wealth effect from the UBI's impact?
Suppose that a subset of lottery winners, driven by behavioral biases like loss aversion, exhibit a tendency to maintain their pre-winning consumption levels, viewing the lottery winnings as "found money" to be saved or invested but not integrated into their regular budget. How would this behavior likely influence the observed effect of wealth on labor earnings in Figure 1, and what econometric technique could be employed to test for this heterogeneous effect?
Suppose that a subset of lottery winners, driven by behavioral biases like loss aversion, exhibit a tendency to maintain their pre-winning consumption levels, viewing the lottery winnings as "found money" to be saved or invested but not integrated into their regular budget. How would this behavior likely influence the observed effect of wealth on labor earnings in Figure 1, and what econometric technique could be employed to test for this heterogeneous effect?
Given a pooled sample of lottery participants comprising PLS, Kombi, Triss-Lumpsum, and Triss-Monthly lottery players, and considering the prevalence of small, frequent wins among PLS participants, how would the Gini coefficient of prize winnings likely change if calculated only for first-time lottery winners across all lottery types, assuming prize amounts remain constant?
Given a pooled sample of lottery participants comprising PLS, Kombi, Triss-Lumpsum, and Triss-Monthly lottery players, and considering the prevalence of small, frequent wins among PLS participants, how would the Gini coefficient of prize winnings likely change if calculated only for first-time lottery winners across all lottery types, assuming prize amounts remain constant?
Assuming the lottery data spans a period during which significant macroeconomic shifts occurred, like fluctuating real interest rates or changes in income tax policies, which econometric technique would be most appropriate to isolate the pure causal effect of lottery winnings on labor supply decisions, controlling for both observed and unobserved heterogeneity?
Assuming the lottery data spans a period during which significant macroeconomic shifts occurred, like fluctuating real interest rates or changes in income tax policies, which econometric technique would be most appropriate to isolate the pure causal effect of lottery winnings on labor supply decisions, controlling for both observed and unobserved heterogeneity?
Given that the dataset identifies 98.7% of lottery participants using name, age, and address, what potential bias might arise from the 1.3% of unidentified participants, and how could this bias affect the estimation of the effect of wealth on labor supply?
Given that the dataset identifies 98.7% of lottery participants using name, age, and address, what potential bias might arise from the 1.3% of unidentified participants, and how could this bias affect the estimation of the effect of wealth on labor supply?
Assuming the marginal propensity to consume (MPC) out of lottery wealth varies substantially across different prize tiers (e.g., individuals winning small prizes exhibit a higher MPC than those winning large prizes), which technique would be most appropriate for precisely estimating the aggregate impact of lottery wealth on aggregate consumption at the population level?
Assuming the marginal propensity to consume (MPC) out of lottery wealth varies substantially across different prize tiers (e.g., individuals winning small prizes exhibit a higher MPC than those winning large prizes), which technique would be most appropriate for precisely estimating the aggregate impact of lottery wealth on aggregate consumption at the population level?
Given that the study uses a discount rate of 1.9% to match the historical real interest rate in Sweden, how would incorporating a time-varying discount rate, reflecting fluctuations in macroeconomic conditions and individual risk preferences, likely impact the estimated long-term effect of lottery wealth on retirement decisions?
Given that the study uses a discount rate of 1.9% to match the historical real interest rate in Sweden, how would incorporating a time-varying discount rate, reflecting fluctuations in macroeconomic conditions and individual risk preferences, likely impact the estimated long-term effect of lottery wealth on retirement decisions?
Assuming that lottery winnings are not perfectly observed due to potential underreporting or strategic concealment by winners, which econometric method would be most appropriate to address this measurement error and obtain consistent estimates of the true effect of wealth on labor supply?
Assuming that lottery winnings are not perfectly observed due to potential underreporting or strategic concealment by winners, which econometric method would be most appropriate to address this measurement error and obtain consistent estimates of the true effect of wealth on labor supply?
Suppose the lottery winnings data are right-censored at a certain high prize level due to confidentiality reasons. How could one estimate the impact of lottery wealth on labor supply while addressing this censoring issue, accounting for potential non-linear effects of wealth?
Suppose the lottery winnings data are right-censored at a certain high prize level due to confidentiality reasons. How could one estimate the impact of lottery wealth on labor supply while addressing this censoring issue, accounting for potential non-linear effects of wealth?
Within the context of the provided regression equation, what is the most precise interpretation of the error term, $\epsilon_{i,0}$, assuming the model is correctly specified?
Within the context of the provided regression equation, what is the most precise interpretation of the error term, $\epsilon_{i,0}$, assuming the model is correctly specified?
If the Nordic country indicator exhibits significant collinearity with the age polynomial in the baseline controls during the test for random assignment, what is the most likely consequence for the regression analysis?
If the Nordic country indicator exhibits significant collinearity with the age polynomial in the baseline controls during the test for random assignment, what is the most likely consequence for the regression analysis?
In assessing the external validity of lottery studies, reweighting population samples to match the age and sex distribution of lottery winners is intended to mitigate which specific form of bias?
In assessing the external validity of lottery studies, reweighting population samples to match the age and sex distribution of lottery winners is intended to mitigate which specific form of bias?
Assuming that lottery prizes are found not to be randomly assigned, and are instead correlated with unobserved determinants of pre-tax annual labor earnings, what econometric strategy would most effectively address the resulting endogeneity?
Assuming that lottery prizes are found not to be randomly assigned, and are instead correlated with unobserved determinants of pre-tax annual labor earnings, what econometric strategy would most effectively address the resulting endogeneity?
If the variance of the error term, $\epsilon_{i,0}$, in equation (1) is found to be heteroskedastic, what is the most appropriate course of action to ensure valid statistical inference?
If the variance of the error term, $\epsilon_{i,0}$, in equation (1) is found to be heteroskedastic, what is the most appropriate course of action to ensure valid statistical inference?
Considering the potential for lottery players to underreport their winnings due to privacy concerns or strategic reasons, how would non-classical measurement error in $L_{i,0}$ likely affect the estimated coefficient on $Z_{i,-1}$ in equation (1)?
Considering the potential for lottery players to underreport their winnings due to privacy concerns or strategic reasons, how would non-classical measurement error in $L_{i,0}$ likely affect the estimated coefficient on $Z_{i,-1}$ in equation (1)?
What is the most compelling reason to include cell fixed effects, $X_{i,0}$, in equation (1) when testing for violations of conditional random assignment?
What is the most compelling reason to include cell fixed effects, $X_{i,0}$, in equation (1) when testing for violations of conditional random assignment?
If the distributions of pre-tax annual labor earnings are highly skewed in both the lottery player sample and the reweighted representative sample, which statistical measure would provide the most robust comparison of central tendency between the two groups?
If the distributions of pre-tax annual labor earnings are highly skewed in both the lottery player sample and the reweighted representative sample, which statistical measure would provide the most robust comparison of central tendency between the two groups?
Assuming that lottery play is habit-forming and that prior lottery losses increase the propensity to participate in subsequent lotteries, what econometric problem is most likely to arise when estimating the effect of lottery winnings on subsequent labor supply using observational data?
Assuming that lottery play is habit-forming and that prior lottery losses increase the propensity to participate in subsequent lotteries, what econometric problem is most likely to arise when estimating the effect of lottery winnings on subsequent labor supply using observational data?
Given the constraint $β_t = β$ imposed on the wealth shock response, and considering the context of analyzing individual-level responses to lottery wealth shocks, what critical assumption must hold true for the validity of subsequent econometric analyses?
Given the constraint $β_t = β$ imposed on the wealth shock response, and considering the context of analyzing individual-level responses to lottery wealth shocks, what critical assumption must hold true for the validity of subsequent econometric analyses?
In the context of analyzing heterogeneous effects by interacting lottery prize amounts ($L_{i,0}$), cell fixed effects ($X_i$), and controls ($Z_{i,-s}$) with a sub-population indicator variable, what specific econometric concern arises if the sub-population exhibits perfect sorting across cells?
In the context of analyzing heterogeneous effects by interacting lottery prize amounts ($L_{i,0}$), cell fixed effects ($X_i$), and controls ($Z_{i,-s}$) with a sub-population indicator variable, what specific econometric concern arises if the sub-population exhibits perfect sorting across cells?
When decomposing the total wealth effect on earnings into extensive- and intensive-margin adjustments related to labor supply, which methodological consideration is paramount to avoid spurious attribution of effects?
When decomposing the total wealth effect on earnings into extensive- and intensive-margin adjustments related to labor supply, which methodological consideration is paramount to avoid spurious attribution of effects?
Given the major tax reform in Sweden during 1990-1991, what econometric strategy would be most effective in mitigating potential biases arising from the joint taxation of capital and labor incomes prior to the reform, when analyzing wealth effects on labor supply?
Given the major tax reform in Sweden during 1990-1991, what econometric strategy would be most effective in mitigating potential biases arising from the joint taxation of capital and labor incomes prior to the reform, when analyzing wealth effects on labor supply?
In the described analysis, all income variables are winsorized at the 0.5th and 99.5th percentiles. What specific econometric concern does this winsorization address, and what potential bias could arise if this step were omitted?
In the described analysis, all income variables are winsorized at the 0.5th and 99.5th percentiles. What specific econometric concern does this winsorization address, and what potential bias could arise if this step were omitted?
Considering the use of population-wide registers originally collected by tax authorities, what source of systematic measurement error is most likely to affect the validity of inferences regarding individual labor supply responses?
Considering the use of population-wide registers originally collected by tax authorities, what source of systematic measurement error is most likely to affect the validity of inferences regarding individual labor supply responses?
What is the most significant threat to internal validity when estimating treatment-effect heterogeneity by interacting lottery winnings with pre-existing individual characteristics?
What is the most significant threat to internal validity when estimating treatment-effect heterogeneity by interacting lottery winnings with pre-existing individual characteristics?
Assuming that lottery winnings are reported net of taxes, and that marginal tax rates vary significantly across the income distribution, what econometric adjustment is most critical when estimating the causal effect of lottery wealth shocks on pre-tax labor income?
Assuming that lottery winnings are reported net of taxes, and that marginal tax rates vary significantly across the income distribution, what econometric adjustment is most critical when estimating the causal effect of lottery wealth shocks on pre-tax labor income?
Suppose a researcher finds statistically significant nonlinear effects of lottery wealth on labor supply. What theoretical explanation would BEST reconcile these findings, assuming standard economic models fail to predict such nonlinearities?
Suppose a researcher finds statistically significant nonlinear effects of lottery wealth on labor supply. What theoretical explanation would BEST reconcile these findings, assuming standard economic models fail to predict such nonlinearities?
Given the potential for lottery winnings to be used for investment in human capital, which analytical approach is MOST suitable for disentangling the direct labor supply effects of wealth from the indirect effects operating through human capital accumulation?
Given the potential for lottery winnings to be used for investment in human capital, which analytical approach is MOST suitable for disentangling the direct labor supply effects of wealth from the indirect effects operating through human capital accumulation?
Consider the econometric specification in equation (2). What potential bias could arise if $L_{i,0}$ is, in fact, not independent of potential outcomes, even conditional on $X_{i,0}$, and how might this compromise the causal interpretation of $\beta_t$?
Consider the econometric specification in equation (2). What potential bias could arise if $L_{i,0}$ is, in fact, not independent of potential outcomes, even conditional on $X_{i,0}$, and how might this compromise the causal interpretation of $\beta_t$?
Given the estimation strategy outlined, what specific econometric challenge arises when estimating the dynamic effects of a wealth shock using OLS, and what methodological refinement could mitigate this issue?
Given the estimation strategy outlined, what specific econometric challenge arises when estimating the dynamic effects of a wealth shock using OLS, and what methodological refinement could mitigate this issue?
In the context of estimating equation (2), what is the most critical assumption required for the validity of Ordinary Least Squares (OLS) estimation, and how might a violation of this assumption manifest in the estimated coefficients?
In the context of estimating equation (2), what is the most critical assumption required for the validity of Ordinary Least Squares (OLS) estimation, and how might a violation of this assumption manifest in the estimated coefficients?
Given the focus on the dynamic effects of wealth, why is controlling for lagged outcomes ($Z_{i,s}$) crucial in equation (2), and what specific econometric problem does it address?
Given the focus on the dynamic effects of wealth, why is controlling for lagged outcomes ($Z_{i,s}$) crucial in equation (2), and what specific econometric problem does it address?
How does clustering standard errors at the individual level impact the statistical inference in this study, and under what specific condition is this approach most appropriate?
How does clustering standard errors at the individual level impact the statistical inference in this study, and under what specific condition is this approach most appropriate?
Why is it crucial to examine pre-lottery years for significant differences in outcomes between those who win large versus small prizes, and what specific threat to internal validity does this address?
Why is it crucial to examine pre-lottery years for significant differences in outcomes between those who win large versus small prizes, and what specific threat to internal validity does this address?
If the analysis reveals significant differences in pre-treatment characteristics despite the expectation of independence conditional on $X_{i,0}$, what alternative identification strategy could be employed to salvage causal inference, and what assumptions would it require?
If the analysis reveals significant differences in pre-treatment characteristics despite the expectation of independence conditional on $X_{i,0}$, what alternative identification strategy could be employed to salvage causal inference, and what assumptions would it require?
What econometric issue arises from including person-year observations for $t = 1, ..., 5$ and baseline controls measured in the year prior to the lottery ($s = -1$) in a modified version of equation (2), and how would you test for it?
What econometric issue arises from including person-year observations for $t = 1, ..., 5$ and baseline controls measured in the year prior to the lottery ($s = -1$) in a modified version of equation (2), and how would you test for it?
Suppose the variance of lottery winnings is significantly different across different lottery types (PLS, Kombi, Triss). How could this heteroskedasticity affect the OLS estimates of (\beta_t), and what method could address this?
Suppose the variance of lottery winnings is significantly different across different lottery types (PLS, Kombi, Triss). How could this heteroskedasticity affect the OLS estimates of (\beta_t), and what method could address this?
In the analysis of lottery winners' labor supply, what are the relative strengths and weaknesses of using cell fixed effects ($X_{i,0}$) versus individual fixed effects, especially considering the time horizon is limited to ten years post-lottery?
In the analysis of lottery winners' labor supply, what are the relative strengths and weaknesses of using cell fixed effects ($X_{i,0}$) versus individual fixed effects, especially considering the time horizon is limited to ten years post-lottery?
Flashcards
Observations Count
Observations Count
The total number of lottery wins observed in the combined dataset.
Unique Individuals
Unique Individuals
The number of unique individuals who won lottery prizes within the observation period identified by name, age, and address.
PLS Lottery
PLS Lottery
A lottery where players win small prizes several times.
Prize Distribution
Prize Distribution
Signup and view all the flashcards
Share of Prizes
Share of Prizes
Signup and view all the flashcards
Kombi Lottery
Kombi Lottery
Signup and view all the flashcards
Triss-Lumpsum Lottery
Triss-Lumpsum Lottery
Signup and view all the flashcards
t = 0
t = 0
Signup and view all the flashcards
yi,t
yi,t
Signup and view all the flashcards
Li,0
Li,0
Signup and view all the flashcards
Xi,0
Xi,0
Signup and view all the flashcards
Zi,s
Zi,s
Signup and view all the flashcards
OLS
OLS
Signup and view all the flashcards
Coefficient estimates
Coefficient estimates
Signup and view all the flashcards
Pre-lottery years
Pre-lottery years
Signup and view all the flashcards
s = -1
s = -1
Signup and view all the flashcards
εi,t
εi,t
Signup and view all the flashcards
Conditional Random Assignment Test
Conditional Random Assignment Test
Signup and view all the flashcards
Cell Fixed Effects
Cell Fixed Effects
Signup and view all the flashcards
Baseline Controls
Baseline Controls
Signup and view all the flashcards
Null Hypothesis (in this context)
Null Hypothesis (in this context)
Signup and view all the flashcards
External Validity
External Validity
Signup and view all the flashcards
Representativeness Concern (Lottery Studies)
Representativeness Concern (Lottery Studies)
Signup and view all the flashcards
Reweighting Samples
Reweighting Samples
Signup and view all the flashcards
Specificity of Wealth Shock
Specificity of Wealth Shock
Signup and view all the flashcards
Wealth Shock
Wealth Shock
Signup and view all the flashcards
Pretax Labor Earnings
Pretax Labor Earnings
Signup and view all the flashcards
Impact of Lottery Wealth
Impact of Lottery Wealth
Signup and view all the flashcards
Estimating Wealth Effect
Estimating Wealth Effect
Signup and view all the flashcards
Earnings Before Winning
Earnings Before Winning
Signup and view all the flashcards
Labor Adjustment Post-Win
Labor Adjustment Post-Win
Signup and view all the flashcards
Restriction βt = β
Restriction βt = β
Signup and view all the flashcards
Five-year Estimates
Five-year Estimates
Signup and view all the flashcards
Heterogeneous Effects
Heterogeneous Effects
Signup and view all the flashcards
Extensive-Margin Adjustments
Extensive-Margin Adjustments
Signup and view all the flashcards
Intensive-Margin Adjustments
Intensive-Margin Adjustments
Signup and view all the flashcards
Winsorizing Income
Winsorizing Income
Signup and view all the flashcards
Time Period: 1991-2010
Time Period: 1991-2010
Signup and view all the flashcards
Population-Wide Registers
Population-Wide Registers
Signup and view all the flashcards
1990-1991 Tax Reform
1990-1991 Tax Reform
Signup and view all the flashcards
Year-2010 SEK
Year-2010 SEK
Signup and view all the flashcards
Study Notes
- The study examines the impact of wealth on labor supply using data from Swedish lottery players.
- Lottery winnings modestly reduce earnings.
- Calibrated dynamic models estimate marginal propensities to earn out of unearned income and labor supply elasticities.
- Earnings response is more significant for winners than their spouses, which does not align with unitary household labor supply models.
Importance of Understanding Labor Supply
- Understanding labor supply responses to wealth changes is crucial for evaluating economic policies like retirement systems and property taxes.
- Accurate estimates of labor supply are valuable for obtaining compensated wage elasticities and informing optimal taxation theory.
Prior Research and Current Study's Approach
- There is limited consensus on the magnitude of wealth's effect on labor supply.
- Challenges exist in isolating exogenous variation in unearned income or wealth for credible wealth-effect estimates.
- This study uses randomized lottery prize assignments to estimate the causal impact of wealth on labor supply.
Advantages of the Study's Lottery Data
- The study effectively controls for the number of lottery tickets bought.
- A large prize pool (approximately US$650 million) enables precise estimation of heterogeneous effects across subsamples.
- Matched to Swedish high-quality administrative data, so labor market outcomes can be studied many years without attrition. External validity concerns about lottery studies are addressed.
Findings on Individual-Level Labor Supply
- Winning a lottery prize immediately reduces earnings.
- Effects are roughly constant over time and last more than ten years.
- Pretax earnings fall by about 1.1% of the prize amount per year.
- A windfall gain of 1 million SEK (about US$140,000) reduces annual earnings by about 11,000 SEK.
- 11,000 SEK is about 5.5% of the sample average.
- Adjustments are primarily made to the number of hours worked.
- There isn't much evidence of heterogeneous or nonlinear effects.
- Winners are not more likely to change employers, industries, or occupations.
- Winning a lottery prize reduces self-employment income, which doesn't line up with previous studies that suggest wealth shocks increase transition into self-employment
Dynamic Labor Supply Model and Elasticities
- A simple dynamic labor supply model with a binding retirement age is used to estimate lifetime marginal propensities to earn (MPE) out of lottery wealth.
- Lifetime MPE varies with age, strongest in the youngest winners (-0.15 to -0.17).
- The average uncompensated labor supply elasticity is close to zero.
- The individual-level compensated (Hicksian) elasticity is 0.10.
- The intertemporal (Frisch) elasticity is 0.14.
- These three estimates are on the lower range of previously reported estimates.
Household-Level Analyses
- Taking into account the labor supply of nonwinning spouses increases shows there's a reduction in the estimated labor supply response by 23%.
- The winner reacts more strongly than the spouse.
- The above is inconsistent with unitary household labor supply models, thus observed labor supply responses should not depend on the identity of the lottery winner.
- The findings support the winner adjusting labor supply more strongly than spouses.
Lottery Samples
Prize-Linked Savings Accounts (PLS)
- Sample of Swedish individuals between 1986 and 2003.
- PLS accounts include a lottery element by randomly awarding prizes to some accounts rather than paying interest.
Kombi Lottery
- Sample of half a million individuals who participated in a monthly ticket-subscription lottery series.
- Lottery tickets to proceeds got to Swedish Social Democratic Party for the postwar era. Subscribers choose their desired number of subscription tickets and billed monthly by the organization.
Triss Lotteries
- Triss is a scratch-ticket lottery run by gaming operator Svenska Spel since 1986
- Included are Triss-Lumpsum and Triss-Monthly prize winners.
- Winners of either type of prize are invited to participate in a morning TV show
Sample Construction
- Merging the three lotteries gives a sample of 435,966 observations, correspoding to 334,532 unique individuals.
- Observations are primarily from PLS lottery (people win small prizes several times).
- The authors exclude individuals who died the same year, lack basic socioeconomic characteristics in public records, or have no income, leaving us with a sample of 426,598 observations.
- Limit to players between 21 and 64 at the time of the win, which reduces the sample to 249,402 observations.
- They drop cells without variation in the amount won, that gives them an estimation sample of 247,275 observations (200,937 individuals.)
Prize Distribution
- All lottery prizes are net of taxes and expressed as units of year-2010 SEK.
- In total more than 5,500 prizes are > 100,000 SEK (US$14,000).
- Almost 1,500 prizes are > 1 million SEK (US$140,000).
- The total prize amount in pooled sample = 4.662 million SEK (about US$650 million).
- PLS and Triss-Monthly each account for 36% of total prize amount, Triss-Lumpsum for 21% and Kombi 7%.
Internal and External Validity
- "Key to identification strategy is that the variation in amount won within cells is random"
- Characteristics determined before lottery should not have any prediction power on the amount won conditional on cell fixed effects.
- The authors reweight representative samples to match the age and sex distribution of the lottery winners.
Estimation Strategy
- The dependent variable is individual i's year-end outcome of interest measured at time t = 0, 1, ..., 10.
Measures and Labor
- All analyses are limited to labor supply outcomes from 1991 until 2010.
- All income variables are winsorized at the 0.5th and 99.5th percentile (2010 SEK currency).
Effect on Annual Earnings
- Primary earnings measure is pretax labor earnings, from three income sources: annual wage earnings, income from self-employment, and income support due to parental leave or sickness absence.
- The effect of lottery wealth is near-immediate, modest in size, and quite stable over time.
Margins of Adjustment
- 1 million SEK windfall decreases participation probability by around 2 percentage points in the five years after
Heterogeneous and Nonlinear Effects
- The effect is similar across lotteries, but may be different than Triss-Monthly.
- If winners have a bias to the present/cannot borrow, they would choose lump-sum, so initial results suggest the opposite
Dynamic Labor Supply Model
- Model is discrete-time and assumes perfect foresight, no uncertainty, and no liquidity constraints for the agent.
Model Simulation
- Simulate to match after-tax results at the individual-level
-
of life years depends on the age at which the winner is awarded the lottery
Simulation Results
- The implied average annual hours are close to the average annual hours in the lottery sample (1,656 hours versus 1,633 hours
- Elasticity is computed for someone who wins at age 50
Wealth
- Examine what the effect of the lottery is on wealth for spouses to married couples
- Results indicate at a given the winner they had on hand wealth that can be correlated between both groups
- Data does provide an estimation of when looking specifically at at TV show about married couples and there are more instances where people buy tickets together
Implied Wealth
- Model can compute the lifetime MPE
Labor
- Elasticity is computed for someone who wins at age 50
- They find there is not a very strong effect with compensation
- They estimate that the impact of wealth based on the number of weekly hours based on salary - approximately $1 millions SEK and that correlates to 4% of a 7 day work period where as the commission has no real difference in the equation.
of hours worked
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.
