How much should we trust regression-kink-design estimates?

In a regression kink (RK) design with a finite sample, a confounding smooth nonlinear relationship between an assignment variable and an outcome variable around a threshold can be spuriously picked up as a kink and results in a biased estimate. In order to investigate how well RK designs handle such confounding nonlinearity, I firstly implement Monte Carlo (MC) simulations and then study the effect of fiscal equalization grants on local expenditure in Japan using an RK design. Results in both the MC simulations and the empirical application suggest that RK estimation without covariates can be easily biased, and this problem can be mitigated by adding observed covariates to the regressors. On the other hand, a smaller bandwidth or a higher-order polynomial, even a quadratic polynomial, tends to result in imprecise estimates although they may be able to reduce estimation bias. In sum, RK estimation with confounding nonlinearity often suffers from bias or imprecision, and estimates are credible only when relevant covariates are controlled for. I also examine how placebo RK estimation can effectively address these issues.     

How much should we trust life satisfaction data? Evidence from the Life in Transition Survey

We analyse responses to two similar life satisfaction questions asked at different points in the same wave of a major cross-country household survey covering the transition region, Turkey and five Western European countries. We show that while the answers to the two questions are broadly consistent for most people, the responses for some groups differ significantly. Respondents of a lower socio-economic status and with a more favourable parental background show systematically higher levels of self-reported satisfaction in the later question. We also find evidence that responses to the later question are influenced by preceding questions on social capital. Our results have important implications for the design and length of household surveys that contain subjective questions.     

How should we collect data on the nursing workforce?

R esearchers who study the nursing workforce use data about nurses and data about nursing jobs. Diifferent approaches - sample surveys and relicensure surveys - used to collect data on RNs, and the benefits and weaknesses of these approaches, are reviewed. Predicting the future supply of nurses is more complex than estimating the current supply, as factors that affect future supply must be considered. Simple forecasts can be created with relatively few variables, which could be collected through relicensure surveys. However, more detailed assessments of potential retirement patterns, changing family norms regarding childcare and household work, and movements into higher education require the rich data more often found in sample surveys.     

How much does schooling disutility matter?

Studies on structural education choice models are often inconsistent in choosing whether and how to include a disutility of education, especially in an environment with risk and wealth inequality. We show that adding a disutility term to the education decision, a human capital investment option, is equivalent to assuming a relationship between wealth, risk, and education. Utility gain from education is increasing in the riskiness of future consumption. A riskier environment further propels an agent to choose the human capital investment option that maximizes future income. If the degree of risk increases heterogeneously across multiple human capital investment options, risk aversion and the precautionary savings motive can compound or negate each other depending on which option has a greater increase in risk.     

How much are electric vehicles driven?

The prospect for electric vehicles as a climate change solution hinges on their ability to reduce gasoline consumption. But this depends on how many miles electric vehicles are driven and on how many miles would have otherwise been driven in gasoline-powered vehicles. Using newly-available U.S. nationally representative data, this paper finds that electric vehicles are driven considerably fewer miles per year on average than gasoline-powered vehicles. The difference is highly statistically significant and holds for both all-electric and plug-in hybrid vehicles, for both single- and multiple-vehicle households, and both inside and outside California. The paper discusses potential explanations and policy implications. Overall, the evidence suggests that today’s electric vehicles imply smaller environmental benefits than previously believed.     

Estimating the demand for settlement balances in the Canadian Large Value Transfer System: How much is too much?

This paper applies a static model of an interest rate corridor to the Canadian data and estimates the aggregate demand for central bank settlement balances in the Large Value Transfer System (LVTS). The analysis takes into account the downward divergence of the overnight interest rate from the target rate, which has been persistent since 2005. The results suggest that a target of $3,000,000,000 for LVTS settlement balances does not seem excessive during the time period when Canadian monetary policy was operating at the effective lower bound (ELB). In addition, the estimation shows that the point elasticity of overnight interest rate is around 0.17 in the ELB framework.     

How should Engel's law be formulated?

Engel's law expresses a “negative stochastic association” of income and the proportion of income that is spent on food. However, there are many quite different notions of “negative stochastic association” and consequently there are different ways of defining Engel's law. We relate these different concepts to Engel's original statistical analysis and show that one must give credit to Engel for the first non-parametric statistical analysis of budget-data.     

Efficient payments: How much do they cost for the Central Bank?

Previous works related to optimal denominations for coins and banknotes consider that the “principle of least effort” that defines an efficient payment is the most important criterion for two main reasons. Firstly, it is more convenient for transactors and, secondly, it limits the production costs of denominations incurred by the central bank. Exploiting production cost data for the U.S. currency system in 2010, we show using simulations that efficient payments actually increase the annual production costs of the Federal Reserve by $156 million. As a consequence, we raise a larger issue for central banks which consists in issuing an efficient denominational mix that is more convenient for transactors and that reduces the production costs of denominations.