Chapter 1 analyzes effects of tax-favored savings plans on savings and retirement decisions in a realistically specified life-cycle model. Individuals face mortality risk and stochastic earnings, allocate assets between conventional savings accounts (CSAs) and tax-deferred accounts (TDAs), make endogenous choice of labor supply and retirement, and make a separate decision on claiming Social Security. The simulations reveal that there is a functional division to some degree between CSAs and TDAs, with the former serving mainly for liquidity and the latter for retirement and bequests. There is tremendous heterogeneity. The tax incentives are generally effective in stimulating new savings for the middle and upper income groups. The higher rate of return on TDAs facilitates wealth accumulation, which consequently and perhaps unintentionally encourages early retirement. Impatient and low-income individuals tend to retire and claim Social Security early. They derive less benefit from TDAs since they face lower marginal tax rates and they have limited resources to take advantage of TDAs. For them, the income effect dominates and TDAs fail to induce new savings.

Chapter 2 attempts quantitatively to measure the efficiency of public spending in developing countries. The efficiency is defined as the distance between observed input-output combinations and an efficiency frontier. Both input- and outputefficiencies are estimated for several health and education output indicators by means of the Free Disposable Hull (FDH) and Data Envelopment Analysis (DEA) techniques. This chapter further seeks to verify empirical regularities associated with cross-country efficiency variation. The panel Tobit regressions reveal that countries are more likely to register lower efficiency if they are faced with higher government expenditure levels, larger wage shares in government budget composition, higher ratios of public to private financing in service provision (health), more prevalence of HIV/AIDS epidemic (health), stronger external aid dependency, and/or higher income inequality (education). Though no causality may be inferred from these exercises, they help point at different factors to understand why some countries spend more resources than others to achieve similar educational and health outcomes.