Archives 2014-2015

Hiver 2015

Vendredi 16 janvier 2015

KABACA, Serdar - Banque du Canada

International Spillovers of Large-Scale Asset Purchases


This paper evaluates the international spillover eff ects of large-scale asset purchases (LSAPs) using a two-country dynamic stochastic general equilibrium model with nominal and real rigidities, and portfolio balance e ffects. Portfolio balance eff ects arise from imperfect substitution between short and long-term bond portfolios in each country, as well as between domestic and foreign bonds within these portfolios. We show that LSAPs lower both domestic and foreign long-term yields, and stimulate economic activity in both countries. International spillover e ffects become larger as the steady-state share of long-term U.S. bond holdings increases in the ROW portfolio, as the elasticity of substitution between short and long-term bonds decreases, or as the elasticity of substitution between domestic and foreign bonds increases. We also fi nd that U.S. asset purchases that generate the same output eff ect as U.S. conventional monetary policy have larger international spillover e ffects. This is because portfolio balance e ffects appear to be stronger under unconventional policy, and foreigners' U.S. bond holdings are heavily weighted towards long-term bonds.

Vendredi 30 janvier 2015

KHAN, Hashmat - Carleton University

House Prices, Consumption, and Government Spending Shocks


We highlight that a broad class of DSGE models with housing and collateralized borrowing predict a fall in both house prices and consumption following positive government spending shocks. By contrast, we show that house prices and consumption in the U.S. rise persistently after identified positive government spending shocks, using a structural vector autoregression methodology and accounting for anticipated effects. We clarify that modifying preferences alone, as previously suggested in the literature, does not help in obtaining the correct house price response. We then show that only when monetary policy strongly accommodates government spending shocks, the impact effects on house prices and total consumption are positive. The model, however, does not deliver the persistent rise in house prices and consumption as evident in the data. Properly accounting for the empirical evidence on government spending shocks and house prices using a DSGE model, therefore, remains a significant challenge.

Mercredi 25 février 2015

NOSTBAKKEN, Linda - Norwegian School of Economics (NHH)

From Fossil Fuels to Renewables: The Role of Electricity Storage


The increased focus on renewable energy sources has sparked considerable interest in developing better energy storage solutions. Since renewable energy sources such as wind and solar are highly variable and unpredictable, significant use of renewables in the energy mix requires better storage solutions. We analyze the determinants of innovation in electricity storage, and study the role of storage in increasing the share of renewable energy. We propose a theoretical model of production and innovation in the electricity sector, where endogenous energy storage innovations affect the relative competitiveness between clean and dirty electricity sources. Next, we empirically test the predictions of our model using a unique global firm-level dataset of electricity patents. Our results suggest that electricity storage plays an important role for the direction of technological innovation in electricity generation (clean vs dirty), and should be considered separately from clean technologies to gain a complete understanding of the incentive structure.

Vendredi 13 mars 2015

DINKELMAN, Taryn - Dartmouth College (ANNULÉ)

Vendredi 27 mars 2015

BARRINGTON-LEIGH, Christopher - McGill University

Rising life satisfaction in Korea: does growth ever get the credit?


South Korea has recently undergone rapid economic and social changes.  The Korean Labor & Income Panel Study (KLIPS) provides a  rich dataset for understanding societal outcomes, which may serve as a model for other developing countries as they join the ranks of the wealthy.  Indeed, South Korea constitutes one of only a few reported cases of a population with a well-measured, dramatic, and sustained rise in average life satisfaction over many years.  However, changes are so rapid in Korea that age and cohort effects are particularly confounded.

In this work I make extensive use of nonparametric estimates in order to disentangle life-course changes from those in the prevailing economic and social conditions.  In canonical models of survey-reported life satisfaction, a polynomial in age is used to account for intrinsic life-cycle effects. Typically, both raw and  residual modeled life satisfaction follow a U-shape with age. In South Korean society, by contrast, there is a dramatic monotonic decline in life satisfaction with age. I begin by showing that every cohort is actually experiencing a dramatic rise in life satisfaction over time, and I discuss the implications for our conception of the determinants of subjective well-being.

By semi-parametrically modelling changes in individual life satisfaction without any parametric dependence on age, I am able to account for most of the overall change in life satisfaction over fourteen years.  To shed light on the various strong assumptions made in this and other typical analyses of panel survey life  satisfaction data, I appeal to the domain satisfaction responses in  KLIPS.

Analysis of overlapping estimates of domain satisfaction from successive cohorts helps in some cases to separate out population-wide development changes from life cycle effects.  I provide evidence that the improvements in overall life quality come in part through improved leisure and improved social relations  outside the family.

I next estimate the relationship between life satisfaction and domain satisfactions directly, arguing that this is not subject to the missing variables problem outlined by Bertrand and Mullainathan (2001). I find strong consistency across age/cohorts using both levels and changes of subjective variables. The findings suggest a strong role for social relations in explaining improvements in life satisfaction, surpassing growth and other factors.

I conclude with more conventional cross-sectional and panel estimates of life satisfaction, in which age enters explicitly. The three approaches prove to be complementary.  I leverage the rich household financial data in KLIPS, and allow for comparison effects in order to address the question of a rat race economy in the midst of the rapid affluence growth which appears to be a major factor in the recent SWL increase. Again finding consistent estimates across subgroups and specifications, I find that the predicted effect of simultaneous, region-wide uniform boosts to income, assets, and consumption would be to reduce SWL, both in a step change and in long term levels. In light of the other evidence, this suggests a strong role for national positive externalities and public goods.

Vendredi 10 avril 2015

FLORES, Nicolas - University of Colorado, Boulder

Public Goods Referenda without Perfectly Correlated Prices and Quantities


This paper examines the incentive properties of probabilistic referenda. In contrast to earlier research in which prices and quantities are perfectly correlated, we consider uncertain and potentially different outcomes for prices and quantities. We provide a theoretical analysis on incentive compatibility and an induced-value experimental test of this theory to gain new insights. First, using a standard design, our results confirm previous findings. Second, our results suggest that moving away from a perfectly correlated design undermines the incentive compatibility result found in other studies. Third, our experimental results are consistent with choices made by risk-averse agents in our theoretical analysis. Our findings would be important for survey design in practice as well as theoretical aspect of CV referenda.

Automne 2014

Mercredi 5 novembre 2014

CORTES, Matias - University of Manchester

The Micro and Macro of Disappearing Routine Jobs: A Flows Approach


The U.S. labor market has become increasingly polarized since the 1980s, with the share of employment in middle-wage occupations shrinking over time. This job polarization process has been associated with the disappearance of per capita employment in occupations focused on routine tasks. We use matched individual-level data from the CPS to study labor market flows into and out of routine occupations and determine how this disappearance has played out at the "micro" and "macro" levels. At the macro level, we determine which changes in transition rates account for the disappearance of routine employment since the 1980s. We find that changes in three transition rate categories are of primary importance: (i) that from unemployment to employment in routine occupations, (ii) that from labor force non-participation to routine employment, and (iii) that from routine employment to non-participation. At the micro level, we study how these transition rates have changed since job polarization, and the extent to which these changes are accounted for by changes in demographic composition or changes in the behavior of individuals with particular demographic characteristics. We find that the preponderance of changes is due to the propensity of individuals to make such transitions, and relatively little due to demographics. Moreover, we find that changes in the transition propensities of the young are of primary importance in accounting for the fall in routine employment.

Vendredi 21 novembre 2014

AKER, Jenny - Tufts University

Call Me Educated: Evidence from a Mobile Monitoring Experiment in Niger


In many rural areas of developing countries, education programs are often implemented through community teachers.  While teachers are a crucial part of the education production function, observing their effort remains a challenge for governments and NGOs.  This paper tests whether a simple monitoring system, implemented via the mobile phone, can improve student learning as part of an adult education program.  Using a randomized control trial in 163 villages in Niger, we randomly assigned half of the villages to a mobile phone monitoring program, whereby teachers, students and the village chief were called on a weekly basis.  There was no incentive component to the program.  The program dramatically affected student performance:  During the first year of the program, reading and math test scores were .15-.25 s.d. higher in monitoring villages than in non-monitoring villages, with relatively stronger effects in the region where monitoring was weakest and for teachers for whom the outside option was lowest.

Vendredi 5 décembre 2014

PARENT, Daniel - HEC Montréal

Performance Pay, the Gender Earnings Gap and Parental Status


We show that the large gender earnings gap at the top of the distribution (the glass ceiling) and the motherhood penalty are associated with each other and both are uniquely associated with performance pay. The US gender earnings gap in performance pay jobs exceeds that in non-performance pay jobs. Yet, this larger gap is driven exclusively by the comparison between mothers and fathers with the unexplained gender earnings gap among parents in performance pay jobs exceeding 50 percent at the top of the earnings distribution. Mothers earn substantially less than childless females in performance pay jobs while fathers earn substantially more than childless males in performance pay jobs. These differences by parental status are muted or completely absent in non-performance pay jobs. We confirm sorting patterns associated with these differentials and suggest alternative theoretical motivations for the observed patterns.