We provide the first, in experimental economics, consistent estimates of a dynamic learning model with a continuous outcome. The econometric approach we propose can be used in many experimental studies including auctions, bargaining with transfers, and gift exchange experiments. We focus on affiliated private value auctions, where subjects are generally assumed to converge to the rule-of-thumb bidding, but our general approach is applicable to many other settings. Our IV estimates suggest that subjects become significantly less aggressive over time; specifically, they decrease their bids in proportion to the previous period’s signal minus bid. However, the inconsistent OLS and FE estimators imply that subjects become significantly more aggressive over time—they raise their bids in proportion to the previous period’s signal minus bid. Our instruments are randomly generated by the experiment, and pass popular weak instrument tests.

This paper specifies the panel data experimental design condition under which ordinary least squares, fixed effects, and random effects estimators yield identical estimates of treatment effects. This condition is relevant to the large body of laboratory experimental research that generates panel data. Although the point estimates and the true standard errors of the estimated average treatment effects are identical across the three estimators, the estimated standard errors differ. A standard F test as well as asymptotic reasoning guide the choice of which estimated standard errors are the appropriate ones to use for statistical inference.

Relying upon an original (country-sector-year) measure of robotic capital ($RK$), we investigate the degree of complementarity/substitutability between robots and workers at different skill levels. We employ nonparametric methods to estimate elasticity of substitution patterns between $RK$ and skilled/unskilled labor over the period 1995–2009. We show that: i) on average, $RK$ exhibits less substitutability with skilled workers compared to unskilled workers, indicating a phenomenon of “RK-Skill complementarity”. This pattern holds in a global context characterized by significant heterogeneity; ii) the dynamic of “RK-Skill complementarity” has increased since the early 2000s; iii) the observed strengthening is more prominent in OECD countries, as opposed to non-OECD countries, and in the Manufacturing sector, compared to non-Manufacturing industries.

Our paper sheds light on Sanitary and Phytosanitary (SPS) cooperation among trading countries. We contribute to the existing literature a data-driven analysis on the effectiveness of various forms (in monetary value, duration, and diversification) of SPS related technical assistance received by 33 countries from 1993 to 2015. The World Trade Organization's (WTO's) SPS Agreement encourages biosecurity for countries through technical assistance, to safeguard human health and productivity from contamination by biological hazards (pests, pathogens, or invasive species). Our panel model finds that WTO's SPS program encourages simultaneously agricultural trade and biosecurity. We implement a Multiple Indicator Solution (MIS) to correct bias from the endogenous technical assistance. The effectiveness of technical assistance depends on geography and the level of development among the heterogeneous countries referred to in our data. This investment in biosecurity benefits both donors and recipients of technical assistance. Based on our results donors should be encouraged to invest in countries with below average resources and abilities.

Much historical yield-monitor data is from fields where a uniform rate of nitrogen was applied. A new approach is proposed using this data to get site-specific nitrogen recommendations. Bayesian methods are used to estimate a linear plateau model where only the plateau is spatially varying. The model is then illustrated by using it to make site-specific nitrogen recommendations for corn production in Mississippi. The in-sample recommendations generated by this approach return an estimated $9/acre on the example field. The long-term goal is to combine this information with other information such as remote sensing measurements.

This article combines cross-national statistical analysis and in-depth historical case studies of Argentina and Chile to explore the relationship between two crucial dimensions of state capacity. We show that information capacity contributes to the development of fiscal capacity. When states have accurate information about their subject populations, territories, and economies, they are more effective at mobilizing revenues. In developing this argument this article makes three broader contributions. First, while existing scholarship either treats distinct dimensions of state capacity as separate entities, or simply assumes that they complement each other, our findings urge scholars to treat state development as sequential and to further investigate how multiple dimensions of state capacity are interrelated. Second, the paper suggests a broader underlying set of mechanisms – economies of scope – which connect these dimensions, and explores them in the specific context of how information capacity facilitates fiscal capacity. Third, we join the scholarship on the importance of societal compliance in the creation of the fiscal state, but with a focus on elite cooperation with the state's information collection efforts, which we show to be crucial to tax state development.

We examine the effect of corruption control on efficiency and its implications for efficiency spillovers by a stochastic frontier model. Our dataset covers 102 countries from 1996 to 2014. We find a positive relationship between corruption control and efficiency. If neighboring countries have difficulty in handling corruption, the country would be negatively affected by its neighbors' corruption through efficiency spillovers. We then compare the efficiency differences across countries for three time periods: 1996–2002, 2002–2008, and 2008–2014. On average, technical efficiencies slightly increased in the second period compared to the first period. In the third period, the efficiencies declined, particularly in China.

This study analyses firms’ labour demand when employers have at least some monopsony power. It is argued that without taking into account (quasi-)monopsonistic structures of the labour market, wrong predictions are made about the effects of minimum wages. Using switching fractional panel probit regressions with German establishment data, I find that slightly more than 80% of establishments exercise some degree of monopsony power in their demand for low-skilled workers. The outcome suggests that a 1% increase in payments for low-skilled workers would, in these firms, increase employment for this group by 1.12%, while firms without monopsony power reduce the number of low-skilled, by about 1.63% for the same increase in remuneration. The study can probably also be used to explain the limited employment effects of the introduction of a statutory minimum wage in Germany and thus leads to a better understanding of the labour market for low-skilled workers.

Inflation rates and their convergence within Euro area have been a major concern, since well before the advent of the single currency. Inflation differentials are a normal phenomenon in any monetary union and even in long-established monetary unions. The aim of this research is to examine the main factors of inflation differentials in the Euro-zone for the period 1999–2018. Our empirical estimates appear to suggest that a one-percentage-point increase in the positive output gap typically leads to an increase of about 20 basis points in the inflation rate of EMU countries. We also find three structural breaks, in 2004, 2008 and in 2010. Since the monetary policy of the European Central Bank is geared at maintaining low and stable inflation, the productivity growth should be increased, and the real effective exchange rates should be decreased and become more homogeneous among EMU. Therefore, countries’ inflation differentials may become less persistent.

We investigate the impact of five types of subsidies granted under the European Union Common Agricultural Policy on the persistent and transient inefficiency of Polish dairy farms. Our research shows that coupled and environmental subsidies reduce transient technical inefficiency, while the opposite is true for Less Favoured Areas (LFA) and other rural subsidies. Simultaneously, environmental, LFA, and other rural subsidies increase persistent technical inefficiency. These results imply that the impact of each type of subsidy on technical efficiency can be different and that the effect of the particular type of subsidy can vary between transient and persistent technical inefficiency.

Wine investment returns can come from overall market trends or price increases with age. Because of the short wine price histories available, market and maturation effects are difficult to separate. Consequently, researchers often obtain dramatically different estimates of investment returns. We find that data sample bias may be the hidden cause of the disparate estimates. In wine auction data, the sample bias refers to a shift in the distribution of which wines are traded as a function of their age. Such sample bias in panel data sampled across many different wine labels can distort the estimation of price increases versus age and consequently impact the estimation of market trends. This analysis shows that segmenting the analysis such that the data panels contain wine labels with similar trading characteristics can lead to a more stable estimation.

The analysis here looks at data from Bordeaux, Italy, Australia, and California. An Age-Period-Cohort (APC) analysis is applied to data panels from each region. Then the data in each region is segmented by a measure of popularity in order to reduce sampling bias. Data thus segmented is then re-analyzed to demonstrate the difference in estimating price appreciation lifecycles and market trends.

This study investigates the determinants of coffee prices received by growers in Costa Rica, paying attention to the impact of environmental, regional, quality, and international aspects in a panel data set for the period 2008–2016. We identify three groups of variables that affect domestic coffee prices. Some of them are external to the control of the coffee growers, such as the international price of green coffee or the power of multinationals; others, such as the altitude where the coffee is harvested or the berries' yield, are related to coffee quality but difficult to modify by coffee growers. The focus of our study is on the third group, which refers to differentiation strategies related to environmental certifications. More specifically, we consider two particularly relevant certifications, which are Fairtrade mills and organic coffee. We find that organic coffee berries received higher prices, but Fairtrade mills report lower average prices than other, non-certified, buyers.

The relationship between temperature and agriculture outcomes in Brazil has been widely explored, overlooking the fact that most of the country's labor force is employed in non-agriculture sectors. We use monthly individual-level panel data spanning the period from January 2015 to December 2016 to ask whether temperature shocks impact non-agriculture wages in formal labor markets. Our results show that additional days in a month that fall within high-temperature ranges have significant adverse effects on real wages. Assuming a uniform climate change scenario where the daily temperature distribution shifts by 2$^{\circ }$C, we calculate income losses for formal workers in non-agriculture markets equivalent to 0.12 per cent of 2015 GDP.

A breeding female’s perceived value is a complicated process and depends on a combination of expected production costs, reproductive success, and calf values. A conceptual asset value model based on female characteristics as signals and net implicit marginal value expectations is developed. A hedonic model based on sequentially sold individuals at multiple Mississippi auction locations is estimated by panel regression. Among other findings, pregnant females are discounted in proportion to abortion risk, which decreases toward birth. A follow-up cost/benefit analysis indicates producers are better off from at home pregnancy checking and selling only nonpregnant females or cow/calf pairs.

Modern economic theory gives an important role to expectations as an influence on outcomes. This paper reviews evidence on how well measures of expectations conform to outcomes. It confirms earlier results that measures taken from financial markets perform poorly as predictors of outcomes. Looking at the individual responses to the Confederation of British Industry’s Industrial Trends Survey, it does find, however, that there are significant correlations between expected and realised outcomes of wages, prices, costs orders and employment. It also finds some evidence that actual prices reflect expected future prices, but with a coefficient much lower than economic theory predicts. There is evidence that forecast errors are explained by past forecasts, as well as revisions to the economic outlook, casting doubt on the idea that firms’ forecasts make the best use of the information available at the time. The paper concludes by observing that, while expectations are undoubtedly important, economists need to build on work looking at how they are derived instead of simply assuming they are rational.

It has become standard practice in the non-life insurance industry to employ generalized linear models (GLMs) for insurance pricing. However, these GLMs traditionally work only with a priori characteristics of policyholders, while nowadays we increasingly have a posteriori information of individual customers available across multiple product categories. In this paper, we therefore develop a framework to capture this a posteriori information over several product lines using a dynamic claim score. More specifically, we extend the bonus-malus-panel model of Boucher and Inoussa (2014) and Boucher and Pigeon (2018) to include claim scores from other product categories and to allow for nonlinear effects of these scores. The application of the proposed multi-product framework to a Dutch property and casualty insurance portfolio shows that customers’ individual claims experience can have a significant impact on the risk classification. Moreover, it indicates that considerably more profits can be gained by accounting for their multi-product claims experience.

This paper reports on the availability of regional capital stock data,1 in the form of new/updated regional (NUTS2 level) capital stock estimates,2 building on an approach (Perpetual Inventory Method) which had been previously developed for the European Commission. The particular focus here is on the UK and how these data are used to shed light on regional labour productivity disparities. Using a NUTS2 level dataset constructed for the period 2000–16, we use a dynamic spatial panel approach from Baltagi et al. (2019) to estimate a model relating productivity to output (growth or levels) and augmented by explicit incorporation of capital stock plus various other covariates such as human capital. We find that regional variations in capital stocks per worker make a significant contribution to regional variations in labour productivity, but the geography of human capital is also highly relevant. Moreover, we give evidence to show that as human capital rises, notably as we move from the regions to London, the impact of capital stock per worker is less. The effect of capital stock depends on the level of human capital.

The generalized linear model (GLM) is a statistical model which has been widely used in actuarial practices, especially for insurance ratemaking. Due to the inherent longitudinality of property and casualty insurance claim datasets, there have been some trials of incorporating unobserved heterogeneity of each policyholder from the repeated observations. To achieve this goal, random effects models have been proposed, but theoretical discussions of the methods to test the presence of random effects in GLM framework are still scarce. In this article, the concept of Bregman divergence is explored, which has some good properties for statistical modeling and can be connected to diverse model selection diagnostics as in Goh and Dey [(2014) Journal of Multivariate Analysis, 124, 371–383]. We can apply model diagnostics derived from the Bregman divergence for testing robustness of a chosen prior by the modeler to possible misspecification of prior distribution both on the naive model, which assumes that random effects follow a point mass distribution as its prior distribution, and the proposed model, which assumes a continuous prior density of random effects. This approach provides insurance companies a concrete framework for testing the presence of nonconstant random effects in both claim frequency and severity and furthermore appropriate hierarchical model which can explain both observed and unobserved heterogeneity of the policyholders for insurance ratemaking. Both models are calibrated using a claim dataset from the Wisconsin Local Government Property Insurance Fund which includes both observed claim counts and amounts from a portfolio of policyholders.

This paper provides a toolbox for the credibility analysis of frequency risks, with allowance for the seniority of claims and of risk exposure. We use Poisson models with dynamic and second-order stationary random effects that ensure nonnegative credibilities per period. We specify classes of autocovariance functions that are compatible with positive random effects and that entail nonnegative credibilities regardless of the risk exposure. Random effects with nonnegative generalized partial autocorrelations are shown to imply nonnegative credibilities. This holds for ARFIMA(0, d, 0) models. The AR(p) time series that ensure nonnegative credibilities are specified from their precision matrices. The compatibility of these semiparametric models with log-Gaussian random effects is verified. Gaussian sequences with ARFIMA(0, d, 0) specifications, which are then exponentiated entrywise, provide positive random effects that also imply nonnegative credibilities. Dynamic random effects applied to Poisson distributions are retained as products of two uncorrelated and positive components: the first is time-invariant, whereas the autocovariance function of the second vanishes at infinity and ensures nonnegative credibilities. The limit credibility is related to the three levels for the length of the memory in the random effects. The limit credibility is less than one in the short memory case, and a formula is provided.