Working Papers

Authors: Ying-Ying Lee

Jan 2015

We study the role of the propensity scores in estimating treatment effects for the treated with a multi-valued treatment.  Assume assignment to one of the multiple treatments is random given observed characteristics.  Valid causal comparisons for the subpopulation who has been treated a particular treatment level are based on two propensity scores - one for the treatment level and one for the counterfactual level.  In contrast to the binary treatment case, these two propensity scores do not add up to one.  This is the key feature that allows us to distinguish different roles of the propensity scores and to provide new insight in well-known paradoxes in the binary treatment effect and missing data literature.  We formally show that knowledge of the propensity score for the treated level decreases the semiparametric efficiency bound, regardless of knowledge of the propensity score for the counterfactual level.  We propose efficient kernel regression estimators that project on a nonparametrically estimated propensity score for the counterfactual level and the true propensity score for the treated level.  A surprising result is implied for the binary treatment effect for the treated: when the propensity scores are known, using one estimated propensity score is not efficient.  Our efficient estimator regresses on a normalized propensity score that utilizes the information contained in the nonparametrically estimated and the true propensity scores.

JEL Codes: C01, C14, C21

Keywords: propensity score, multi-valued treatment, semiparametric efficiency bound, unconfoundedness, generated regressor

Reference: 738

Individual View

Authors: Alan Beggs

Jan 2015

This paper studies learning in monotone Bayesian games with one-dimensional types and finitely many actions. Players switch between actions at a set of thresholds.  A learning algorithm under which players adjust their strategies in the direction of better ones using payoffs received at similar signals to their current thresholds is examined.  Convergence to equilibrium is shown in the case of supermodular games and potential games.

JEL Codes: C72, D83

Keywords: bayesian games, monotone strategies, learning, stochastic approximation, supermodular games

Reference: 737

Individual View

Authors: Alexander Naumov

Jan 2015

We propose a simple structural model of the upstream sector in the oil industry to study the determinants of costs with a focus on its relationship with the price of oil. We use the real oil price, data on global drilling activity and costs of drilling to estimate a three-dimensional VAR model. We use short run restrictions to decompose the variation in the data into three structural shocks. We estimate the dynamic effects of these shocks on drilling activity, costs of drilling and the real price of oil. Our main results suggest that (i) a 10% increase (decrease) in the oil price increases (decreases) global drilling activity by 4% and costs of drilling by 2% with a lag of 4 and 6 quarters respectively; (ii) positive shocks to drilling activity affect the oil price negatively; (iii) shocks to costs of drilling do not have a permanent effect on the price of oil.

JEL Codes: Q31

Keywords: Natural Resource Extraction, Crude Oil Price, Upstream Cost

Reference: 152

Individual View

Authors: Rabah Arezki, Valerie A Ramey, Liugang Sheng

Jan 2015

This paper explores the effect of news shocks on the current account and other macroeconomic variables using worldwide giant oil discoveries as a directly observable measure of news shocks about future output ̶ the delay between a discovery and production is on average 4 to 6 years. We first present a two-sector small open economy model in order to predict the responses of macroeconomic aggregates to news of an oil discovery. We then estimate the effects of giant oil discoveries on a large panel of countries. Our empirical estimates are consistent with the predictions of the model. After an oil discovery, the current account and saving rate decline for the first 5 years and then rise sharply during the ensuing years. Investment rises robustly soon after the news arrives, while GDP does not increase until after 5 years. Employment rates fall slightly for a sustained period of time.

JEL Codes: E00, F3, F4

Keywords: news shocks, current account, saving, investment, employment, oil, discovery

Reference: 153

Individual View

Authors: Rebecca K Scott

Dec 2014

A recent surge of literature on tax salience has included studies that use tax type as a proxy for salience.  The relationship between tax type and salience is not always apparent, however, nor is salience the only feature by which taxes differ.  In fact, taxes' behaviour over time suggests an alternative explanation for consumers' tax sensitivity: rational habits or forward-looking investment.  Consumers affected by these intertemporal issues will be more responsive to price components that carry stronger signals about future prices-price components such as the specific taxes posited to be particularly salient.  This paper develops a model to disentangle and test for tax salience and rational habits effects.  Differentiating the two effects is important, as they carry vastly different policy implications: tax salience implies that publicity and nominal incidence matter; rational habits imply all that matters is an instrument's effects on price behaviour.  Examining the case of beer demand, I find evidence that favours a rational habits mechanism over a salience effect.  Examining the case of gasoline demand, I find rational habits to be the more plausible explanation for consumers' sensitivity to specific taxes, though a salience effect cannot be ruled out definitively.

JEL Codes: D12, Q41, H30

Keywords: gasoline demand, tax salience, rational habits, price elasticity

Reference: 736

Individual View

Authors: David Hendry, Jurgen A. Doornik

Dec 2014

Big Data offer potential benefits for statistical modelling, but confront problems like an excess of false positives, mistaking correlations for causes, ignoring sampling biases, and selecting by inappropriate methods.  We consider the many important requirements when searching for a data-based relationship using Big Data, and the possible role of Autometrics in that context.  Paramount considerations include embedding relationships in general initial models, possibly restricting the number of variables to be selected over by non-statistical criteria (the formulation problem), using good quality data on all variables, analyzed with tight significance levels by a powerful selection procedure, retaining available theory insights (the selection problem) while testing for relationships being well specified and invariant to shifts in explanatory variables (the evaluation problem), using a viable approach that resolves the computational problem of immense numbers of possible models.

JEL Codes: C51, C22

Keywords: Big Data, Model Selection, Location Shifts, Autometrics

Reference: 735

Individual View

Authors: Erik Mohlin, Robert Ostling, Joseph Tao-yi Wang

Nov 2014

We exploit a unique opportunity to study how a large population of players in the field learn to play a novel game with a complicated and non-intuitive mixed strategy equilibrium.  We argue that standard models of belief-based learning and reinforcement learning are unable to explain the data, but that a simple model of similarity-based global cumulative imitation can do so.  We corroborate our findings using laboratory data from a scaled-down version of the same game, as well as from three other games.  The theoretical properties of the proposed learning model are studied by means of stochastic approximation.

JEL Codes: C72, C73, L83

Keywords: Learning, imitation, behavioral game theory, evolutionary game theory, stochastic approximation, replicator dynamic, similarity-based reasoning, generalization, mixed equilibrium

Reference: 734

Individual View

Authors: Simon Quinn, Tom Gole

Nov 2014

When members of a committee have incentives to agree with each other, they over-weight public information: this can generate status quo bias.  We test this hypothesis using a novel field experiment -  a large debate tournament with random assignment of judges to committees.  To analyse our experimental data, we develop a new structural methodology for estimating discrete dynamic Bayesian games using Markov Perfect Equililbrium.  Our method allows for correlated unobservable signals and for rational dynamic updating of coordination preferences along the equilibrium path.  Our structural estimates show that judges with greater desire to coordinate are more likely to vote for teams with better past records; this shows that, in a committee context, public information can cause coordination on weaker candidates.

JEL Codes: C57, C93

Keywords: committees, discrete games, identification, field experiments, discrimination

Reference: 733

Individual View

Authors: James Malcomson

Nov 2014

This paper explores the implications of specific training for relational contracts.  A standard result for sustaining a relational contract is that the parties must jointly receive a surplus over what they can get by separating.  This has been interpreted as employees with relational contracts having discretely higher pay and productivity than inherently equally productive, or near equally productive employees without relational contracts.  Investment in specific training relaxes the incentive constraints on relational contracts, so the optimal level of investment can be higher for those with a relational contract than for those without, adding further to the productivity of those employed under a relational contract.  But the additional cost of optimal investment precisely offsets the post-investment surplus for marginal employees in relational contracts, which removes the discontinuity in the joint payoff from a relational contract.  An example shows that with optimal investment there may not even be a discontinuity in productivity between those employed with a relational contract and those employed without one because the incentive constraints on the former result in lower effort despite their higher training.

JEL Codes: C73, D82, D86

Keywords: relational incentive contracts, investment, specific training, dual labour market

Reference: 732

Individual View

Authors: J-F Carpantier, W N Vermeulen

Nov 2014

This paper tests the theoretically founded hypothesis that the surge of SWF establishments is determined by three main factors: 1) the existence of natural resources profits, 2) the government structure and 3) the ability to invest usefully in the domestic economy.  We test this hypothesis on a sample of 20 countries that established an SWF in the period 1998-2008 by comparing them to the roughly 100 countries that did not set up a fund in the same period. We find evidence for all three factors. The results suggest that SWFs tend to be established in countries that run an autocratic regime and have difficulties finding suitable opportunities for domestic investments. We do not find the net foreign asset position of a country to be similarly related to the explanatory variables, indicating that the establishment of an SWF is distinct from a national accounting result. We argue that our results indicate that it is relevant to study how an SWF interacts with the domestic economy and government policy.

JEL Codes: E21, E62, F39, G23, H52

Keywords: Sovereign Wealth Fund, Institutions, natural resources

Reference: 148

Individual View

Authors: Yehuda Levy

Nov 2014

A long-standing open question raised in the seminal paper of Kalai and Lehrer (1993) is whether or not the play of a repeated game, in the rational learning model introduced there, must eventually resemble play of exact equilibria, and not just play of approximate equilibria as demonstrated there.  This paper shows that play may remain distant - in fact, mutually singular - from the play of any equilibrium of the repeated game.  We further show that the same inaccessibility holds in Bayesian games, where the play of a Bayesian equilibrium may continue to remain distant from the play of any equilibrium of the true game.

JEL Codes: C65, C72, C73

Keywords: Rational Learning, Repeated Games, Nash Equilibrium

Reference: 731

Individual View

Authors: Jasper van Dijk

Nov 2014

This paper shows that within a regional economy, employment in the nontradable sector benefits from attracting jobs in the tradable sector.  I rework Moretti's study of U.S. cities (AER 2010) and find that one new job in a given city's tradable sector will result into 1.02 new jobs in the nontradable sector in the same city.  I show Moretti overestimated the size of this local multiplier by 0.57, because he made five perfunctory assumptions that had a major impact on his results.  Subsequently I show that Moretti's assertion that skilled tradable jobs have a larger multiplier than unskilled tradable jobs is not supported by the data.  The evidence provided by Moretti was only significant due to an endogeneity effect.

Keywords: Local labour market, multiplier, tradable, nontradable

Reference: 730

Individual View

Authors: Francesco Zanetti, Haroon Mumtaz

Oct 2014

This paper studies how key labor market stylized facts and the responses of labor market variables to technology shocks vary over the US postwar period.  It uses a benchmark DSGE model enriched with labor market frictions and investment specific technological progress that enables a novel identification scheme based on sign restrictions on a SVAR with time-varying coefficients and stochastic volatility.  Key findings are: i) the volatility in job finding and separation rates has declined over time, while their correlation varies across time; ii) the job finding rate plays an important role for unemployment, and the two series are strongly negatively correlated over the sample period; iii) the magnitude of the response of labor market variables to technology shocks varies across the sample period.

JEL Codes: E32, C32

Keywords: Technology shocks, labor market frictions, Bayesian SVAR methods, sign restrictions

Reference: 728

Individual View

Authors: Mark Armstrong,John Vickers

Oct 2014

We provide a simple necessary and sufficient condition for when a multiproduct demand system can be generated from a discrete choice model with unit demands.

JEL Codes: D01, D11

Keywords: Discrete choice, unit demand, multiproduct demand functions

Reference: 729

Individual View

Authors: Andrew Martinez

Oct 2014

This paper compares annual one-year-ahead and five-year-ahead forecasts from government agencies for the U.S. gross federal debt and deficit from 1984 to 2013.  Other studies have compared two of these agencies' forecasts, but not for debt.  The current paper finds that the forecast from the Analysis of the President's Budget performs best across both horizons but does not encompass the other forecasts.  Instead, each of the forecasts lacks information included by the other agencies and therefore a combination of all three outperforms any forecast individually.

JEL Codes: C53, H68

Keywords: Evaluating Forecasts, Government Forecasting, Macroeconomic Forecasts, Forecast Encompassing, Deficit

Reference: 727

Individual View

Authors: Alexander James

Oct 2014

A surprising feature of resource-rich economies is slow growth. It is often argued that natural-resource production impedes development by creating market or institutional failures. This paper establishes an alternative explanationa slow-growing resource sector.  A declining resource sector is disproportionately reflected in resource-dependent countries. Additionally, there is little evidence that resource dependence impedes growth in non-resource sectors. More generally, this paper illustrates the importance of considering industry composition in cross-country growth regressions.

JEL Codes: Q2; Q3; O1

Keywords: Resource Dependence; Economic Growth; Resource Curse

Reference: 147

Individual View

Authors: H Peyton Young

Oct 2014

Social norms are patterns of behavior that are self-enforcing at the group level: everyone wants to conform when they expect everyone else to conform.  There are multiple mechanisms that sustain social norms, including a desire to coordinate, fear of being sanctioned, signaling membership in the group, or simply following the lead of others.  This article shows how stochastic evolutionary game theory can be used to study the dynamics of norms.  We illustrate with a variety of examples drawn from economics, sociology, demography, and political science.  These include bargaining norms, norms governing the terms of contracts, norms of retirement, duelling, footbinding, medical treatment, and the use of contraceptives.  These cases highlight the challenges of applying the theory to empirical cases.  They also show that the modern theory of norm dynamics yields insights and predictions that go beyond conventional equilibrium analysis.

JEL Codes: C73, A120, O10

Keywords: evolutionary game theory, equilibrium selection, stochastic stability

Reference: 726

Individual View

Authors: Ansgar Walther

Sep 2014

Banks create excessive systemic risk through leverage and maturity mismatch, as financial constraints introduce welfare-reducing pecuniary externalities.  Macroprudential regulators can achieve efficiency with simple linear constraints on banks' balance sheets, which require less information than Pigouvian taxes.  These can be implemented using the Liquidity Coverage and Net Stable Funding ratios of Basel III.  When bank failures are socially costly, microprudential regulation of leverage is also required.  Optimally, macroprudential policy reacts to changes in systematic risk and credit conditions over the business cycle, while microprudential policy reacts to both systematic and idiosyncratic risk.

JEL Codes: G18, G21, G28, E44

Keywords: Systemic risk, leverage, maturity mismatch, macroprudential regulation, liquidiity, capital requirements, fire sales

Reference: 725

Individual View

Authors: James Forder

Sep 2014

There is a widely believed but entirely mythical story to the effect that the discovery of 'the Phillips curve' was, in the 1960s and perhaps later, an inspiration to inflationist policy.  The point that this is a myth is argued in Forder, Macroeconomics and the Phillips curve myth, OUP 2014.  One aspect of the explanation of how that myth came to be widely believed is considered in this paper.  It is noted that the expression 'Phillips curve' was applied in a number of quite distinct and inconsistent ways, and as a result there was, by about 1980, an enormous confusion as to what that label meant.  This confusion, as well as the multiplicity of possible meanings, it is suggested, made the acceptance of the myth much easier, and is therefore part, although only part, of the story of its acceptance.

JEL Codes: B22, B29

Keywords: Phillips curve, expectations, Phillips curve myth

Reference: 724

Individual View

Authors: James Forder

Sep 2014

Friedman (1968) - his famous Presidential Address to the American Economic Association - contains an elementary error right at the heart of what is usually supposed to be the paper's crucial argument.  That is the argument to the effect that during an inflation, changing expectations shift in Phillips curve.  It is suggested that the fact of this mistake, and of its having gone all-but unnoticed are points of historical interest.  Further reflections, drawing on the arguments of Forder (2014) Macroeconomics and the Phillips curve myth, are suggested.

JEL Codes: B22, E31

Keywords: Phillips, Friedman, Expectations

Reference: 723

Individual View

Authors: Pawel Dziewulski,John Quah

Sep 2014

Suppose we observe a finite number of input decisions made by a firm, as well as the prices at which those inputs were acquired.  What conditions on the set of observations are necessary and sufficient for it to be consistent with a firm choosing inputs to maximize profit, subject to a production function exhibiting production complementarities?  In this paper, we develop an axiomatic characterisation of this hypothesis and also develop a test that can be easily applied to finite data sets.

JEL Codes: D21, D24

Keywords: profit maximisation, production complementarities, supermodular production, modern manufacturing, cyclical monotonicity, quasilinear preferences

Reference: 722

Individual View

Authors: John Knight, Ramani Gunatilaka

Aug 2014

The paper contrasts early theories of the utility function (starting with Bentham and elaborated by Jevons) with the modern theory (laid down by Fisher and Samuelson).  The former include in the utility function not only the sensation of current events but also the memory of past events and the anticipation of future events.  The alternative hypotheses are tested by introducing both past and expected future income into the estimated subjective well-being function, using an appropriate data set for China.  The tests favour the early theories.  Implications are drawn.

JEL Codes: B13, B21, D60

Keywords: Anticipation, China, Discounted utility, Memory, Subjective well-being, Utility function

Reference: 721

Individual View

Authors: Lucas Bretschger, Christos Karydas

Aug 2014

We study the effects of greenhouse gas emissions on optimum growth and climate policy by using an endogenous growth model with polluting non-renewable resources. Climate change harms the capital stock. Our main contribution is to introduce and extensively explore the naturally determined time lag between greenhouse gas emission and the damages due to climate change, which proves to be crucial for the transition of the economy towards its steady state. The social optimum and the optimal abatement policies are fully characterized. The inclusion of a green technology delays optimal resource extraction. The optimal tax rate on emissions is proportional to output. Poor understanding of the emissions diffusion process leads to suboptimal carbon taxes and suboptimal growth and resource extraction.

JEL Codes: Q54, O11, Q52, Q32

Keywords: Non-Renewable Resource Dynamics; Pollution Di usion Lag; Optimum Growth; Clean Energy; Climate Policy

Reference: 144

Individual View

Authors: Antoine Bommier, Lucas Bretschger, Francois Le Grand

Aug 2014

The paper proves the existence of equilibrium in nonrenewable resource markets when extraction costs are non-convex and resource storage is possible.  Inventories atten the consumption path and eliminate price jumps at the end of the extraction period. Market equilibrium becomes then possible, contradicting previous claims from Eswaran, Lewis and Heaps (1983). We distinguish between two types of solutions, one with immediate and one with delayed build-up of inventories. For both cases we do not only characterize potential optimal paths but also show that equilibria actually exist under fairly general conditions. It is found that optimum resource extraction involves increasing quantities over a period of time. What is generally interpreted as an indicator of increasing resource abundance is thus perfectly compatible with constant resource stocks.

JEL Codes: Q30, C62, D92, D41

Keywords: Exhaustible resources, nonconvex extraction cost, equilibrium existence, resource storage

Reference: 146

Individual View

Authors: Lucas Betschger, Alexandra Vinogradova

Aug 2014

Climate physics predicts that the intensity of natural disasters will increase in the future due to climate change. One of the biggest challenges for economic modeling is the inherent uncertainty of climate events, which crucially affects consumption, investment, and abatement decisions. We present a stochastic model of a growing economy where natural disasters are multiple and random, with damages driven by the economy's polluting activity. We provide a closed-form solution and show that the optimal path is characterized by a constant growth rate of consumption and the capital stock until a shock arrives, triggering a downward jump in both variables. Optimum mitigation policy consists of spending a constant fraction of output on emissions abatement. This fraction is an increasing function of the arrival rate, polluting intensity of output, and the damage intensity of emissions. A sharp response of the optimum growth rate and the abatement share to changes in the arrival rate and the damage intensity justifes more stringent climate policies as compared to the expectation-based scenario. We subsequently extend the baseline model by adding climate-induced fluctuations around the growth trend and stock-pollution effects, demonstrating robustness of our results. In a quantitative assessment of our model we show that the optimal abatement expenditure at the global level may represent 0.9% of output, which is equivalent to a tax of $71 per ton carbon.

JEL Codes: O10, Q52, Q54

Keywords: Climate policy, uncertainty, natural disasters, endogenous growth

Reference: 145

Individual View

Loading Papers...