Working Papers

This article explores how Work From Home (WFH) affects workers’ career progression in the post-pandemic context of the United Kingdom, elucidating the mechanisms that drive these outcomes. Using data from the discrete choice experiment fielded between July and December 2022 among 1,000 managers, we show that teleworkers, whether in hybrid or full-time WFH arrangements, face a disadvantageous evaluation by managers compared to their office-based counterparts. The adverse effect of hybrid teleworking is due to the fact that employers consider hybrid workers are less productive than onsite workers. Full-time teleworkers are penalized even if they display the same performance at work as onsite workers. We demonstrate this penalty to be driven by the fact that managers consider full-time teleworkers to be less committed to work than onsite workers. Consistently with past research, we also find that WFH affects workers’ careers differently depending on their gender and parental obligations and that managers’ assumptions about workers’ performance and commitment allow to explain at least some of these differences.

This paper implements automated trading strategies with buy/sell signals based on Autoregressive Integrated Moving Average (ARIMA) and Vector autoregression (VAR) models. ARIMA and VAR models are compared based on several forecast error measures and investment performance statistics. The data used in this thesis are daily closing prices of Dow Jones Industrial Average, NASDAQ Composite and NYSE Composite indices. The trading period covers 20 years of data from 2000-11-30 to 2020-11-30. The sensitivity analysis is made by changing the initial parameters to test how robust the methods are to these changes. Results show that although ARIMA model performed remarkably well during the volatile periods, VAR based strategy had better investment performance and was less robust to the changes compared to the ARIMA based strategy. Additionally, we have found that error metrics might be insufficient to evaluate performance of forecasting models, as VAR with higher forecast errors outperformed ARIMA model in algorithmic trading strategies.

In this paper, we investigate the effect of respondents’ attitudes concerning distributive justice in payments on their stated preferences for programmes reducing ambient air pollution in four cities in Poland. By combining two multi-factorial survey experiments, we propose a novel approach of incorporating justice attitudes into non-market valuation. In the first experiment – a factorial survey experiment (FSE) – we record justice attitudes towards payments. In the second experiment – a choice experiment (CE) – we elicit stated preferences for air pollution reduction programmes. As a modelling framework, we employ a hybrid choice model. The same respondents undertook both experiments in separate surveys one to two weeks apart, minimising the likelihood of biased estimates of the effect of justice attitudes on stated preferences. The results indicate a substantial effect of the justice attitude on the stated willingness to pay. The proposed approach could be used for joint modelling of justice attitudes and preferences in a wide range of fields, contributing further insights into their interactions.

This paper proposes a novel approach to hedging portfolios of risky assets when financial markets are affected by financial turmoils. We introduce a completely novel approach to diversification activity not on the level of single assets but on the level of ensemble algorithmic investment strategies (AIS) built based on the prices of these assets. We employ four types of diverse theoretical models (LSTM - Long Short-Term Memory, ARIMA-GARCH - Autoregressive Integrated Moving Average - Generalized Autoregressive Conditional Heteroskedasticity, momentum, and contrarian) to generate price forecasts, which are then used to produce investment signals in single and complex AIS. In such a way, we are able to verify the diversification potential of different types of investment strategies consisting of various assets (energy commodities, precious metals, cryptocurrencies, or soft commodities) in hedging ensemble AIS built for equity indices (S&P 500 index). Empirical data used in this study cover the period between 2004 and 2022. Our main conclusion is that LSTM-based strategies outperform the other models and that the best diversifier for the AIS built for the S&P 500 index is the AIS built for Bitcoin. Finally, we test the LSTM model for a higher frequency of data (1 hour). We conclude that it outperforms the results obtained using daily data.

This paper provides two possible explanations for the mixed evidence regarding migration by low-skilled workers to generous welfare states. Using a model of unrestricted migration to a developed, destination country, which provides a direct and equal social benefit to all its residents, we study the impact of the benefit in a country on the size of its low-skilled immigrant population under the assumption that migration is driven by an international difference in returns to skills, employment opportunities in the destination country, and by the generosity of the benefit in that country. We find that the social benefit affects the size of the country’s low-skilled immigrant population not only directly, via the difference between the benefit and its cost in the form of taxation, but also via two indirect channels. The benefit incentivizes taking up low-skilled jobs among the destination country’s native residents, which adversely affects wages of low-skilled workers in that country, and it increases the risk of unemployment of low-skilled workers therein. Prospective low-skilled migrants view these side effects of the benefit as “stay away” factors. Simulation of the model based on 2018 data for EU-15 economies without Luxembourg highlights the importance of indirect channels in curtailing the inflow of low-skilled migrants to a generous welfare state. When only direct channels are accounted for, semi-elasticities of the size of the low-skilled immigrant population with respect to the social benefit are between 0.2 and 0.54. When indirect channels are allowed to play their roles, the positive relationship between the social benefit and low-skilled immigration is significantly reduced; the semi-elasticities range from 0.13 to 0.4. At the level of the model’s fundamentals, the variation in semi-elasticities between EU-15 countries is largely explained by differences in the size of the welfare state and in efficiency of the labor market across these countries.

This paper investigates the issue of an adequate loss function in the optimization of machine learning models used in the forecasting of financial time series for the purpose of algorithmic investment strategies (AIS) construction. We propose the Mean Absolute Directional Loss (MADL) function, solving important problems of classical forecast error functions in extracting information from forecasts to create efficient buy/sell signals in algorithmic investment strategies. Finally, based on the data from two different asset classes (cryptocurrencies: Bitcoin and commodities: Crude Oil), we show that the new loss function enables us to select better hyperparameters for the LSTM model and obtain more efficient investment strategies, regarding risk-adjusted return metrics on the out-of-sample data.

In this study, we investigated how performance in a number of puzzles (decisions under risk) depended on the framing. The puzzles, drawn and adapted from existing literature, were designed to expose well-established cognitive biases could lead respondents to select intuitive yet incorrect answer. Subjects were randomly assigned to one of three treatments: a third of the sample saw puzzles framed in terms of COVID-19, another third about a common cold, and the remaining group about unemployment. Across five continents, we collected over 8,000 observations. We found that framing of the puzzles affected performance, prompting questions regarding the external validity of these puzzles. Treatments associated with more severe threats, such as COVID and Unemployment, elicited stronger (negative) emotions compared to the common cold. Moreover, these emotional reactions were also linked to performance, and their levels correlated negatively with the number of correctly solved puzzles.

In this paper, we analyse the trade implications of accession to the European Union (EU) for four Visegrad (V4) countries (Czechia, Hungary, Poland, and Slovakia). The estimation of the effects of integration with the EU was carried out as a comparative case study using the synthetic control method (SCM). The selection of control variables of “donor countries”, necessary to construct counterfactual trade flows of “synthetic” countries, is based on the gravity model. We analyse the impact of the accession to the EU for 307 country pairs, which covers about 85% of the trade flows of V4 countries.  Using this SCM approach we find a positive impact of accession to the EU on the country's performance 15 years after accession. The trade growth resulting from the accession - in comparison to counterfactual aggregates - is large but differentiated. The exports of V4 countries doubled, except for Hungary, while imports increased by 51% on average. Consequently, the trade balances of V4 countries greatly improved compared to counterfactuals. The trade flows strongly increased among V-4 countries, in contrast with the “Hub and spoke” hypothesis. We checked the robustness of our empirical results by running a placebo test.

This paper presents the results of investment strategies based on predictions from an ARIMA with exogenous variables (ARIMAX/ARIMAX-Garch) model, using the prices of selected commodities and companies from the DJIA index as explanatory variables. The explained variables are four Invesco ETF funds (DBE, DBA, DBP, DBB) corresponding to baskets of energy, agricultural, precious, and industrial metals. The models are optimized using the Walk-Forward technique, and the selection of exogenous variables is based on Granger causality tests. By analyzing the results, we conclude that ARIMAX/ARIMAX-Garch models are not useful tools for making buy or sell decisions for the selected commodity baskets. Out of the 80 estimated models, 44 outperform the Buy & Hold strategy, however, none achieved statistically significant results. Combining individual models into an investment portfolio reduced the risk without significantly reducing the profit, enabling us to consistently beat the benchmark. We also observe that using returns of commodities listed on stock exchanges is more effective than using stock returns. Sensitivity analysis shows instability in results with changes in the length of the training and testing windows. The highest annual return rate of 15.37% from 02.01.2008 to 01.12.2022 was characterized by an ARIMAX model with one commodity exogenous variable.

Cultural goods provide numerous non-market benefits to society. Estimates of the benefits are needed for benefit-cost analyses, helping to inform cultural policy decisions and aiming at the efficient allocation of public funds. The non-market benefits cannot be assessed through market transactions. While original non-market valuation studies require substantial budgets and time, a benefit transfer approach offers an alternative. It enables the application of empirical estimates from existing original studies conducted at one site to approximate the value at another site. This study provides the first international benefit transfer for performing arts and examines the reliability of various benefit transfer approaches. We use empirical data from two separate stated preference valuation surveys conducted in Denmark and in Poland. Our results suggest that the benefit function transfer accounting for differences in purchasing power parity between the countries can generate transfer errors as low as 3-6%, indicating high reliability of the transferred values.

This article presents an extensive analysis of the private ownership of cars and other transport modes in Warsaw, Poland, with a focus on understanding mobility patterns and exploring sustainable alternatives to private car usage. It provides a comprehensive description of car ownership trends, highlighting the high and growing number of cars per capita in Poland, particularly in Warsaw. The existing transport system in Warsaw, including the public transport network and related policies, is summarized. A literature review examines institutional, socio-economic, and individual factors influencing mobility behaviors and the dynamics of recent changes in car usage and alternative modes of transport. The analysis identifies barriers and opportunities for the adoption of sustainable mobility solutions, while discussing policy implications at the national and international levels.

We aim to answer the question of whether using forecasted stock returns based on machine learning and time series models in a mean-variance portfolio framework yields better results than relying on historical returns. Nevertheless, the problem of the efficient stock selection has been tested for more than 50 years, the issue of adequate construction of mean-variance portfolio framework and incorporating forecasts of returns in it has not been solved yet. Stock returns portfolios were created using ’raw’ historical returns and forecasted return based on ARIMA-GARCH and the XGBoost models. Two optimization problems were concerned: global maximum information ratio and global mini-mum variance. Then strategies were compared with two benchmarks – an equally weighted portfolio and buy and hold on the DJIA index. Strategies were tested on Dow Jones Industrial Average stocks in the period from 2007-01-01 to 2022-12-31 and daily data was used. The main portfolio performance metrics were information ratio* and information ratio**. The results showed that using forecasted returns we can enhance our portfolio selection based on Markowitz framework, but it is not a universal solution, and we have to control all the parameters and hyperparameters of selected models.

Work from home (WFH) has been a part of the professional landscape for over two decades, yet it was the COVID-19 pandemic that has substantially increased its prevalence. The impact of WFH on careers is rather ambiguous, and a question remains open about how this effect is manifested in the current times considering the recent extensive and widespread use of WFH during the pandemic. In an attempt to answer these questions, this article investigates whether managerial preferences for promotion, salary increase and training allowance depend on employee engagement in WFH. We also explore the heterogeneity of the effects of WFH on careers across different populations by taking into account the employee’s gender, parenthood status, frequency of WFH as well as the prevalence of WFH in the team. An online discrete choice experiment was run on a sample of over 1,000 managers from the United Kingdom. The experiment was conducted between July and December 2022, and thus after the extensive use of this working arrangement during the COVID-19 pandemic. The findings indicate that employees who WFH are less likely to be considered for promotion, salary increase and training than on-site workers. The pay and promotion penalties for WFH are particularly true for men (both fathers and non-fathers) and childless women, but not mothers. We also find that employees operating in teams with a higher prevalence of WFH do not experience negative career effects when working from home. The findings underline the importance of individual factors and familiarisation as well as social acceptance of flexible working arrangements in their impact on careers.

This study compares well-known tools of technical analysis (Moving Average Crossover MAC) with Machine Learning based strategies (LSTM and XGBoost) and Ensembled Machine Learning Strategies (LSTM ensembled with XGBoost and MAC). All models were compared to Buy and Hold benchmark and evaluated using Performance Metrics, that is Annualized Return Compounded, Maximum Drawdown, Maximum Loss Duration, and three types of Information Ratio. This research uses daily S&P 500 index data ranging from 2000 to 2023. Every strategy was optimized with novel walk forward approach consisting of numerous in sample and out of sample periods. MAC and best performing ML methods were subjected to sensitivity analysis. The results show that LSTM ensembled with XGBoost and MAC yields the most promising results in terms of risk-adjusted returns which suggest further research focused on ensembling of individual ML strategies. Finally, we show that classical methods of technical analysis (that is, MAC) are much less robust and indifferent to change in hyperparameters than machine learning based algorithms, especially LSTM.

Algorithmic trading has been a central theme in numerous research papers, combining knowledge from the fields of Finance and Mathematics. This thesis aimed to apply basic Technical Analysis indicators for predicting price movement of three noble metals: Gold, Silver, and Platinum in a form of multi-class classification. That task was performed using four algorithms: Logistic Regression, k-Nearest Neighbors, Random Forest and XGBoost. The study incorporated feature filtering methods such as Kendall-tau filtering and PCA, as well as five different data frequencies: 1, 5, 10, 15 and 20 trading days. From a total of 40 potential models for each metal, the best one was selected and evaluated using data from period 2018-2022. The result revealed that models utilizing only Technical Analysis indicators were able to predict price movements to a significant extent, leading to investment strategies that outperformed the market in two out of three cases.

Research from the US argues that women will benefit from a structural labour market change as the importance of social tasks increases and that of manual tasks declines. This article contributes to this discussion in three ways: (a) by extending the standard framework of task content of occupations in order to account for diversity of social tasks; (b) by developing measures of occupational task content tailored to the European context; and (c) by testing this argument in 13 European countries. Data are analysed from the European Skills, Competences, Qualifications and Occupations Database and the European Structure of Earnings Survey. The analysis demonstrates that relative to men the structural labour market change improves earnings potential of women working in low- and middle-skilled occupations but not those in high-skilled occupations. Women are overrepresented in low paid social tasks (e.g. care) and are paid less for analytical tasks than men.

While research in constitutional economics has made important progress in recent decades, the factors determining whether a constitution is complied with have only received scant attention. We contribute to that narrow literature by studying how personal traits of political leaders are associated with constitutional compliance. Compliance levels of 943 political leaders between 1950 and 2010 can be explained by their education, entry into office, political experience, military background, and whether they are members of extreme left-wing parties. Specifically, under former military officers and Communist leaders, constitutional compliance is significantly lower. The combined effect of these two traits corresponds to the difference between democracy and nondemocracy. This implies a difficult normative question, which we discuss in our conclusion: Should constitutions set entry barriers for high political offices that are based on leader characteristics?

Using data from “Generation and Gender Survey” for Poland, we study the relationship between women’s relative income within the household, as measured by the female share of total household income, and women’s involvement in housework. We find that households in which the woman contributes more to the total household income are more likely to share housework equally. We also find that individual gender norms matter both for women’s involvement in unpaid work at home and for the observed link between the female share of income and inequality between the partners in the division of housework. Women from less traditional households are found to be more likely to share housework equally. However, this negative relationship between the female share of household income and female involvement in housework is not observed among more traditional couples.

Despite their recent global expansion in nominal terms, many transboundary nature protected areas tend to avoid hands-on cross-border co-operation. One common explanation which is widely seen a major obstacle towards the concerted transboundary conservation is international free riding on the centralised decision-making level. I examine empirically whether international free-riding is embedded in citizens’ stated preferences for extended protection in the case of two transboundary nature protected areas Białowieża Forest and Fulufjället. I scrutinise a sub-set of merged survey samples from the four countries involved, including only the citizens assumingly incentivised to free-ride on unilateral foreign country’s conservation action. I apply attitudinal indicators to form effect-coded variables that measure free-riding, and control for use value, nationality, individual socioeconomic characteristics, and incentive compatibility of the survey design. The results indicate no widespread tendency of international free-riding; the conclusion is maintained with varying modelling approaches or sampling strategy employed.

Technological change and globalisation have been transforming the structure of labour demand in favour of workers performing cognitive tasks. Even though past research has found that labour force participation is an important determinant of fertility behaviour, few studies have addressed the fertility effects of the long-term structural changes of labour market. To fill this gap, we measure the cognitive task content of work at the occupation level using data from the Employment Survey of the German Federal Institute for Vocational Education and Training (BiBB). We link this contextual information with employment and fertility histories of women and men from the German Socio-Economic Panel 1984-2018 (GSOEP). With event history models, we find that fertility transitions of men working in occupations characterised by high cognitive task intensity are accelerated. We also observe elevated birth risks among women in occupations requiring cognitive labour. However, this pattern is more ambiguous, as we find that non-working women also experience elevated birth rates.