Research Seminars

We investigate whether Google searches can be used to forecast stock returns. We find that short-term (daily and weekly) increase in search activity leads to negative excess stock returns with subsequent reversal, whereas long-term (quarterly) increase in search activity leads to long-lasting positive returns.  The connection between searches and subsequent returns is much stronger for smaller companies. Finally, we examine a trading strategy based on our findings and notice that our strategy outperforms a passive strategy even after taking transaction costs into account.

In most OECD-countries, immigrants have lower employment and higher unemployment than natives. On average, the gap in labor market outcomes is larger in countries with more immigrant friendly attitudes. To explain this pattern, this paper developes a theory based on labor market institutions and welfare state institutions. In countries where labor market institutions give native workers more influence, and where the social safety net is more generous, native workers face less direct wage competition from immigration. As a result, the general population is more immigrant friendly and income inequality is dampened, but empoyment among immigrants suffer thwarting the potential economic benefits from immigration (and in some cases immigration becomes a financial burden for the public sector). The theory is confirmed using data for 21–28 OECD countries using OLS-regressions and Bayesian model averaging over all 512 theoretically possible model specifications to cope with the model selection problem which is particularly severe in small samples where the number of suggested explanations is high. The unemployment gap is bigger in countries where collective bargaining agreements cover a larger share of the labor market, and the employment gap is bigger in countries with more generous social safety nets. The education of immigrants and migrant integration policies have no explanatory value. Finally, the prediction that countries with smaller labor market gaps have higher income inequality is also supported by the data.

A recent debate in the forecasting literature revolves around the inability of macroeconometric models to improve on simple univariate predictors, since the onset of the so called Great Moderation. This paper explores the consequences of equilibrium indeterminacy for quantitative forecasting through standard reduced form forecast models. Exploiting U.S. data on both the Great Moderation and the preceding era, we first present evidence that (i) higher (absolute) forecastability obtains in the former rather than the latter period for all models considered, and that (ii) the decline in volatility and persistence captured by a finite-order VAR system across the two samples need not be associated with inferior (absolute or relative) predictive accuracy. Then, using a small-scale New Keynesian monetary DSGE model as laboratory, we generate artificial datasets under either equilibrium regime and investigate numerically whether (relative) forecastability is improved in the presence of indeterminacy. It is argued that forecasting under indeterminacy with e.g. unrestricted VAR models entails misspecification issues that are generally more severe than those one typically faces under determinacy. Irrespective of the occurrence of non-fundamental (sunspot) noise, for certain values of the arbitrary parameters governing solution multiplicity, the pseudo out-of-sample VAR-based forecasts of inflation and output growth can outperform simple univariate predictors. For other values of these parameters, by contrast, the opposite occurs. In general, it is not possible to establish a one-to-one relationship between indeterminacy and superior forecastability, even when sunspot shocks play no role in generating the data. Overall, our analysis points towards a 'good luck in bad policy' explanation of the (relative) higher forecastability of macroeconometric models prior to the Great Moderation period.

Presentation adds to the research on the role of the housing sector in transition countries. The East European Housing Model (Hegedüs and Tosics, 1996) summarizes the main characteristics and elements of the housing system in the centrally planned economy, which was a social-economic system with high job security, low – highly subsidized – housing costs, and small income differences. The crucial question is how East European Housing Model has changed after the transition. Is it possible to define a post-socialist housing model or is the emerging new housing model similar to one of existing housing models in the developed world?

The Czech Republic is a country with a strong attachment of women to the labor market but one of the longest paid parental leave durations. Using a difference-in-differences methodology, we study the effect of two reforms of duration of parental allowance on the labor market status of mothers 2-7 years after childbirth. While the 1995 reform prolonged parental allowance from 3 to 4 years and the 2008 reform introduced a flexible schedule that allowed shortening of leave from 4 to 2 or 3 years, both reforms kept the job protection period at 3 years, allowing us to study the impact of monetary incentives setting aside changes in job security. We find that the 1995 reform prolonged the parental leave of at least one third of mothers and shifted the post-leave unemployment spell to the time when a child turns 4, while the 2008 reform achieved a reversal of the impact of the 1995 reform, but only to a lesser extent.

Previous research has established that good-looking political candidates win more votes. We extend this line of research by examining differences between parties on the left and on the right of the political spectrum. Our study combines data on personal votes in real elections with a web survey in which 2,513 non-Finnish respondents evaluated the facial appearance of 1,357 Finnish political candidates. We find that political candidates on the right are better looking in both municipal and parliamentary elections and that they have a larger beauty premium in municipal, but not in parliamentary, elections. As municipal candidates are relatively unknown, the beauty-premium gap indicates that voters - especially those to the right - use beauty as a cue for candidate ideology or quality in the municipal elections.

We examine the relationship between information and price dispersion in the retail gasoline market. We first show that the clearinghouse models in the spirit of Stahl (1989) generate an inverted-U relationship between information and price dispersion. Past empirical studies of this relationship have relied on (intertemporal) variation in internet usage and adoption to measure the number of consumers that have access to the clearinghouse. We construct a new measure of information based commuter data from Austria. Regular commuters can freely sample gasoline prices on their commuting route, giving us spatial variation in the share of informed consumers. We use detailed information on gas station level price to construct various measures of price dispersion. Our empirical estimates of the relationship are in line with the theoretical predictions.

The Czech population is one of the most equal societies in terms of households' disposable incomes and has the lowest level of relative poverty in Europe. We ask if the Czech tax and benefit system helps to achieve this low inequality and poverty. We test this hypothesis with the best available data on households from the Czech Statistical Office - the Survey of Income and Living Conditions (SILC) for direct taxes and social benefits combined with the Household Budget Survey (HBS) for indirect taxes. We thus combine detailed data on household's income and expenditure for the first time. We show that market income, especially due to the inclusion of pensions, is quite egalitarian. We find that the narrowly defined tax-benefit system (direct taxes and social benefits) actually does not change the poverty rate, while the indirect taxes increase it. The Czech tax and benefit system thus does not seem to contribute to the one of the world's lowest levels of relative poverty and inequality. We further provide the first estimates of the redistributive effectiveness of a number of social and tax policies. Among other findings, we show that aid in material need benefits are the most effective in decreasing poverty gap and income inequality, while the child allowance is the largest benefit in terms of coverage of poor individuals.

We report the results of a simple statistical choice task based on independent and identically distributed (iid) variables. In our experiment subjects were asked to bet on the future performance of players in a coin toss task. Subjects exhibit a strong, irrational bias towards placing their bets on players with a good guessing history in the coin toss task. We also show that the result cannot be attributed to confusion induced by the BDM mechanism. Subjects' elicited preferences are compatible with prescriptive luck beliefs. That is, the idea that luck is a somehow deterministic and personal attribute.

The talk will explore quadratic hedging of derivatives in a complete and incomplete financial market. Starting with a discrete-time setting we will provide several numerical illustrations one of which leads to the famous Black-Scholes formula in the continuous-time limit . The talk is based on joint work with Jan Kallsen and Stephan Denkl of CAU Kiel.

This paper uses a laboratory experiment to analyse the effect that transactions costs and inertia have on charitable giving. We conjecture that transaction costs are likely to have a greater effect on donations if the solicitations are received when potential donors are busy (when the opportunity cost of time is high) as opposed to when they have time on their hands to donate (when the opportunity cost of time is low). If donations do not have to be made immediately, inertia could also become a factor as people might intend to give, but then postpone making the payment until they have more time, and having postponed making the donation once, keep doing so until the opportunity to donate has passed. We find that introducing a transaction cost to a standard Dictator Game with charity as a recipient, and manipulating whether the donation can be made when we know subjects have time on their hands, reduces donations providing evidence of a transaction cost effect. Some weak evidence is found that giving people more time to donate reduces donations, which is consistent with an inertia effect.

We investigate the extent to which lottery financing of public goods crowds out voluntary contributions driven by social preferences. On average, we find presence of such crowding out effect. We then classify subjects by the strength of their conditional cooperation and find that the extent of crowding out is increasing with the the level of conditional cooperation, especially under a higher lottery prize. We interpret this finding as crowding-out of voluntary contributions driven by reciprocity.

We investigate the trip scheduling behavior of Dutch train commuters, using data collected during a large-scale reward experiment with more than 1000 participants. Railway subscription owners were invited to join an experiment where they could earn (distance-based) monetary rewards if they traveled outside the (morning and evening) peak hours. A dedicated smartphone app was used to observe the trip timing and routing behavior of the participants. Compared to the pre-measurement, the share of peak trips decreased by ca. 20% during the reward period, and by 9% during the post-measurement. We estimate departure time choice models, drawing from multiple data sources to compute the attribute values for all choice alternatives. We are able to derive plausible revealed preference (RP) values for the participants' willingness-to-pay for reducing travel time, schedule delays, the number of transfers and crowdedness, and for increasing reliability.

We present a methodology for detecting manipulation of public procurement and evidence showing how policies that create discontinuous incentives to avoid transparent competition lead to manipulation and active waste by procurement officials. Our methodology exploits a natural experiment in which new discretionary thresholds in the anticipated value of procurements were established. Manipulations reveal through bunching of procurements below the new thresholds and affect 11% of relevant contracts. Manipulations lead to increases in the chance of allocating contracts to anonymously owned firms often related to corrupt behavior, increases in the final prices of procurements and preferential prices for anonymous contractors.