Public Finance Field Exam: August 2009 - Economic Analysis of Homelessness and Labor Taxes, Exams of Economics

Download Public Finance Field Exam: August 2009 - Economic Analysis of Homelessness and Labor Taxes and more Exams Economics in PDF only on Docsity! Department of Economics Graduate Office Public Finance Field Exam August 2009 Directions: Answer all questions. There is no choice, although you may answer the questions in whatever order you prefer. 1. In-kind vs. Cash Redistribution The City of San Francisco has an estimated 2,800 homeless adults. In November 2002, 59 percent of voters in the City of San Francisco approved Proposition N to reform public services provided to the city’s homeless population. In place of cash welfare payments, Proposition N provides treatment for drug and alcohol addiction, mental care, housing and job training. For a typical homeless person, the law would reduce the cash welfare payment from almost $360 to $60 per month, but guarantee housing, health care and job services. a. For simplicity in this question, let us assume that the reform reduces individual cash transfers from $360 to $60 and provides instead housing services worth $300 a month in the market (ignore the health care and training aspects for now). Consider the choice between housing and other goods, and discuss the impact of Proposition N on a "typical" homeless person’s budget constraint. What is the expected impact on housing and non-housing expenditures? Be precise about the set of assumptions you make. b. For simplicity in this question, let us assume that the reform reduces cash transfers from $360 to $60 and provides one free meal everyday for the homeless (worth $300 per month or $10 per meal at market prices) [forget about housing, health care, or training benefits here]. To get the meal, the person simply has to show up and queue at a given location to get the free meal. Assume that queuing for the daily meal is one hour on average. Nobody checks that people queuing to get the free meals are truly homeless. Suppose the homeless person can get $3 per hour on average panhandling. Explain how this program impacts the budget constraint of the homeless. c. Continuing with the scenario in b., suppose now that the meal delivery is made more efficient so that it takes only 10 minutes of queuing (instead of one hour) to get the meal. Are the homeless better-off? Explain how the incentives for the non-homeless to take-up the free meals change. Suppose the government has a fixed budget for meals so that the value of individual meals is inversely proportional to take-up. From the perspective of the homeless, what is the optimum queuing time? (no formula required, simply explain the arguments). d. Proponents of Proposition N cite a "recent study from the New England Journal of Medicine [that] showed a direct correlation between cash grants like those given in San Francisco and drug overdoses." Presumably, the intended interpretation is that providing cash welfare leads to overuse of drugs. You have no funds to run a randomized experiment; how would you evaluate empirically the effect of cash assistance on drug use using the San Francisco 2002 reform? Discuss the data you would need, the empirical methods and any major issues in the interpretation of your results. Public Finance Field Exam, August 2009 page 2 e. Suppose the new program provides training services to the homeless. Economists have demonstrated that such training programs are effective: the present discounted value of future earnings increases by more than the cost of training. Assume also that such training services are provided (at cost) by the private market. Should the government subsidize such training programs for the homeless? (discuss the arguments in favor and against, stating precisely the situations where they apply). 2. Labor Supply and Taxes Let us assume (for simplicity) that the current income tax has the following simple structure. Married couples can deduct $30,000 from their incomes, and singles (single or widowed adults or single parents) can deduct $15,000. Income less this deduction is called taxable income. Taxable income faces a marginal tax rate of 20% up to $100,000, and of 30% above $100,000. a. Explain why such an income tax system generates a marriage penalty, and how it can potentially discourage labor supply of secondary earners. Suppose that the government decides to eliminate the marriage penalty by taxing all incomes at the individual level according to the tax schedule for the singles described above (that is, married individuals would be taxed separately according to their own income). b. Assuming no behavioral responses to this tax change, describe which couples would loose and which couples would gain from the reform. Do you think that the new tax system will be more or less redistributive that the previous one? Is the government likely to gain or to loose tax revenue from this reform (assuming again no behavioral responses)? [be as precise as you can about the statistics you would need to answer quantitatively] c. Suppose now that the reform affects labor supply but not marriage decisions. Explain which labor supply responses this reform is likely to generate. Suppose you have cross-sectional data on family structure and incomes (for each member of the couple for married households) before and after the reform, and that male labor supply is not affected by the reform. Describe a simple differences-in-differences methodology that would allow you to estimate the labor supply elasticity of married women with respect to the net-of-tax rate (one minus the marginal tax rate). What are the potential biases of the method you just described and how would you check the validity of the method if more data were available? d. Suppose you now want to distinguish between the extensive labor supply responses (participation in the labor market) and the intensive labor supply responses (level of earnings on the job). Explain how cross-sectional data (before and after the reform as in question c. could allow you to estimate the extensive elasticity of labor supply of married women. Explain why, in the presence of extensive labor supply responses, cross-sectional data does not allow the estimation of intensive labor supply responses.