Programme organizer: Balázs Reizer
Gergő Tóth, Johannes Wachs, Riccardo Di Clemente, Ákos Jakobi, Bence Ságvári, János Kertész
and Balázs Lengyel
Ride-share matching algorithms generate income inequality
Eszter Bokányi and Anikó Hannák
Abstract
Despite the potential of online sharing economy platforms such as Uber, Lyft, or Foodora to
democratize the labor market, these services are often accused of fostering unfair working conditions and low wages. These problems have been recognized by researchers and regulators but the size and complexity of these socio-technical systems, combined with the lack of transparency about algorithmic practices, makes it difficult to understand system dynamics and large-scale behavior. This paper combines approaches from complex systems and algorithmic fairness to investigate the effect of algorithm design decisions on wage inequality in ride-hailing markets. We first present a computational model that includes conditions about locations of drivers and passengers, traffic, the layout of the city, and the algorithm that matches requests with drivers.
We calibrate the model with parameters derived from empirical data. Our simulations show that
small changes in the system parameters can cause large deviations in the income distributions
of drivers, leading to a highly unpredictable system which often distributes vastly different incomes to identically performing drivers. As suggested by recent studies about feedback loops in
algorithmic systems, these initial income differences can result in enforced and long-term wage
gaps.
Kiss Hubert János: Coopetition in group contest
https://old.kti.krtk.hu/wp-content/uploads/2019/05/MTDP1911.pdf
26/09/2019 Mihály LAKI (tba)
IE Seminar – Győző GYÖNGYÖSI (Kiel Institute) 03/10
Győző GYÖNGYÖSI (Kiel Institute for the World Economy)
Financial distress and student performance
This paper studies the effect of household financial distress on students’ cognitive skill development. We use household foreign currency credit expansion with a large and unexpected depreciation of the domestic currency. The depreciation increased the debt burden of households borrowing in foreign currencies but not of households borrowing in the local currency. We measure household foreign currency debt exposure at the postal code level, and student achievement using administrative student level standardized test scores. Our empirical strategy compares the students’ cognitive skills attending the same class but living in different postal codes. Preliminary results show that a 10 percent unexpected local debt shock decreases the math and reading skills by .045 standard deviations. We also explore the channels through which the crisis affected student development.
Arieda Muco (CEU)
Campaign Financing Ban and (Un)Intended Consequences
To limit the power of corporations in politics, in 2015, the Brazilian Supreme Court banned corporate donations to politicians campaigns. Using a novel dataset of political connections, we document that after the ban firms continue to contribute to politicians’ campaigns, via their managers, exploiting loopholes in the law. Moreover, manager contributions after the ban are comparable to firm contributions before the ban. Finally, managers themselves are more likely to enter the political race. These findings imply several unanticipated consequences that should be taken into consideration for policy evaluation.
Róbert SOMOGYI: Prioritization vs Zero-rating: Discrimination on the Internet
Abstract:
This paper analyzes two business practices on the mobile internet market, paid prioritization and zero-rating. Both violate the principle of net neutrality by allowing the internet service provider to discriminate different content types. In recent years these practices have attracted considerable media attention and regulatory interest. The EU, and until recently the US have banned paid prioritization but tolerated zero-rating under conditions. With prioritization, the ISP delivers content at different speeds and it is equivalent to a discrimination in terms of quality. With zero-rating, the ISP charges different prices for content and it is equivalent to a discrimination in terms of prices. We first show that neither of these practices lead to the exclusion of a content provider, a serious concern of net neutrality advocates. The ISP chooses prioritization when traffic is highly valuable for content providers and congestion is severe, and zero-rating in all other cases. Furthermore, investment in network capacity is suboptimal in the case of prioritization and socially optimal under zero-rating.
(Department of Social Science, UCL Institute of Education)
Is ‘first in family’ a good indicator for widening university participation?
This paper asks whether ‘first in family’ (FiF) to attend university is a good marker of disadvantage in the context of the Widening Participation (WP) agenda in the UK. Currently, 15 of the 24 Russell Group universities use FiF as an indicator for Widening Participation, but very little is known about this indicator’s properties and how it overlaps with other indicators of disadvantage. We use Next Steps (formerly the Longitudinal Study of Young People in England, LSYPE) linked to administrative data, the National Pupil Database (NPD), to provide the first comprehensive analysis of the FiF measure and whether or not it captures additional disadvantage over and above other markers of socioeconomic disadvantage. We employ probability and classification models to look at the relative predictive power of being FiF compared to the other commonly used WP measures to predict university participation and graduation. Our preliminary results show that school-level factors are extremely important in predicting higher education success and the FiF measure does capture some additional disadvantage beyond traditional WP indicators. This research has policy implications for universities as they choose indicators for disadvantage in their admissions processes.
István KÓNYA, Judit KREKÓ, Gábor OBLATH
Labor shares in the EU sectoral effects and the role of relative prices
The paper studies the labor share among countries of the European Union, with a particular attention to newer member states of Central and Eastern Europe (CEEU). After discussing methodological issue s in the computation of the labor share, we present various stylized facts at the country level, and also for broad sectors within the aggregate economy. We find that CEEU countries typically have lower labor shares, both in the aggregate and at the sector all level. Structural change, while quite pronounced among the CEEU economies, plays only a minor role in the evolution of the labor share. The exception is agriculture, which for some countries have a sizable impact on the level and dynamics of the labor share – partly because of important measurement problems. We also document links between productivity, the relative prices of consumption and investment, and the labor share. In particular, we find that a significant part of the difference in conventionally measured labor shares between the more developed EU countries and less developed CEEU countries can be attributed to differences in relative prices. We discuss possible explanations, and show that given reasonable assumptions, a simple two – sector model is able to account for the main findings.
Álmos TELEGDY – Balázs MURAKÖZY: Placebó vagy teljesítményfokozó? Az Európai Unió vállalati támogatásainak hatásai a cégek viselkedésére
10/01/2019 @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K13-14.
Abstract: TBA
Márta BISZTRAY: A külföldi működőtőke beruházások és a nemzetközi kereskedelemben való részvétel közvetett hatásai
21/01/2019 MONDAY @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K11-12.
Abstract: TBA
Kinga MARCZELL (CEU) The Effect of Managers’ Health Shocks on Employment Practices with Gergely HAJDU (CEU)
17/01/2019 @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K11-12.
We investigate the effect of managers’ health shocks on the separation rate of their employees. Our hypothesis is that previous illness experience of division leaders at a company may affect their attitudes towards employees. To test our hypothesis, we measure changes in the separation rate of employees assigned to managers before and after the managers’ illness episodes. Our results show that employee separation rate increases in the manager’s employee pool after the manager’s illness by 8%, a phenomenon mostly driven by an increase in the number of dismissals, as opposed to an increase in the number of employees leaving the firm voluntarily. We provide a descriptive analysis of managers’ own employment outcomes as well. We find that adverse employment effects are present even four years after the illness episode. While 18.23% of previously ill managers has no job by this time, the corresponding ratio is only 13.02% for the control group, and this difference is almost entirely coming from the difference in their likelihood of having a manager position. Conditional on staying at the firm, managers’ wage decreases by 13.4% following the year of illness, compared to the wage evolution of their matched healthy counterparts.
László CSATÓ (BCE, MTA-SZTAKI): A universal university ranking from the revealed preferences of the applicants
31/01/2019 @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K11-12.
https://arxiv.org/abs/1810.04087
A methodology is presented to rank universities on the basis of the applicants’ revealed preferences. We exploit a crucial feature of the centralised admissions system to higher education in Hungary: a student is admitted to the first programme where the score-limit is achieved. This makes it possible to derive a partial preference order of each applicant. Our approach integrates the information from all students participating in the system, does not require any preliminary selection of criteria, and is essentially independent of ad hoc weights, while it is able to reflect even non-measurable college characteristics. The suggested procedure is implemented for ranking faculties in the Hungarian higher education between 2001 and 2016. We demonstrate that the ranking given by the least squares method has favourable theoretical properties such as size invariance and bridge player independence, is robust with respect to the aggregation of preferences, and performs well in practice.
Gábor BÉKÉS, Márta BISZTRAY: Területi egyensúly – a munkaerőpiac és az ingatlanárak kapcsolata Magyarországon
21/02/2019 @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K11-12.
Panka BENCSIK, Universitiy of Sussex
Stress on the sidewalk: The mental health costs of close proximity crime
29/11/2018 @ MTA Humán Tudományok Kutatóháza, 1097 Budapest, Tóth Kálmán utca 4. K13-14.
Using individual level administrative panel data from Hungary between 2003-2011, we analyse the causal effect of job loss on disability pension utilisation on a five-year horizon. We look at individuals being displaced due to a mass layoff, and compare their labour force status to non-laid-off individuals with similar employment and health history, who were chosen with propensity score matching.
According to our estimates from discrete time hazard models, being laid off increases the transition probability to disability retirement by 60–90% (or 1-1.5 %points in 2-4 years). We find greater than average effect among women, the higher educated and those who were in worse health before.
Outpatient, inpatient and pharmaceutical expenditures increase to 3.5-4 times when a laid-off individual takes up disability pension, and decrease slightly afterwards, but do not reach the pre-disability levels. The time pattern of health expenditures around disability retirement is similar to those individuals who were not laid off but still became disabled. Overall, disability retirement of laid-off workers seems to be mostly the consequence of health shocks.
Abstract:
Koren Miklós
Expatriate Managers and Firm Performance
08/11/2018
Abstract:
Better managed firms are more productive. So are foreign owned firms. Using exhaustive administrative data from Hungary 1992–2014, we study the performance of firms led by expatriate managers. Studying more than 2,000 foreign acquisitions, we compare firms that bring in foreign management to those that remain locally managed. Foreign managed firms increase their productivity faster. Larger, more capital intensive and more productive firms are more likely to receive a foreign manager. Owners from distant countries and tax havens are more likely to leave domestic management in place. Our study can help understand the modes of foreign investment and provide a link between foreign ownership and management practices.
04/10/2018
Gergely HAJDU (CEU)
Motivated Beliefs and Prosociality
27/09/2018
Gábor KERTESI
The effect of neonatal intensive care system on reducing infant mortality (Evidence from 45 years of establishing and operating a national network)
Co-authors: Tamás HAJDU, Gábor KÉZDI, Miklós SZABÓ (SOTE), Ágnes SZABÓ-MORVAI
20/09/2018
László GADÁR – János ABONYI: Graph configuration model based evaluation of the education-occupation match
To study education&occupation matchings we developed a bipartite network model of education to work transition and a graph configuration model based metric. We studied the career paths of 15 thousand Hungarian students based on the integrated database of the National Tax Administration, the National Health Insurance Fund, and the higher education information system of the Hungarian Government. A brief analysis of gender pay gap and the spatial distribution of over-education is presented to demonstrate the background of the research and the resulted open dataset. We highlighted the hierarchical and clustered structure of the career paths based on the multi-resolution analysis of the graph modularity. The results of the cluster analysis can support policymakers to fine-tune the fragmented program structure of higher education.
The details of this research are published in PLOS ONE:
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0192427
All the files and the R code are available at:
https://github.com/abonyilab/Edu_Mine_Graph
13/09/2018
Biao Xiang
/School of Anthropology, University of Oxford/
The changing political economy of migrant evictions in China
Abstract
In the wake of a fire incident that killed 19 migrants in Beijing in November 2017, the municipality government evicted possibly hundreds of thousands migrants, leaving them temporarily homeless. Although migrant evictions had occurred periodically, this round of ousting represents something new. Previous evictions had been a means used by local government to increase land value through cyclical demolition. The removal in 2017 is however associated with the agenda of ending this pattern of accumulation-through-demolition and distancing local government from business. This eviction campaign reveals new contradictions brought about by the radical recentralization of state power that started around 2013. By examining how different social groups reacted to the eviction, this paper also seeks to rethink the notions of justice and power based on people’s lived experiences.
24 May 2018
Koen Declerq (KU Leuven)
Gender differences in applying for STEM programs in higher education: evidence from a policy shift in Hungary (with Júlia Varga)
Abstract
We study how admission policies differently affect enrollment decisions of men and women and how these policies can increase enrollment of women in STEM programs. We apply our analysis to Hungary, where 44% of men and only 11% of women preferred a STEM program in their first year in higher education. We investigate how a policy reform that limited access to subsidized non-STEM programs, differently affected application decisions of men and women. We find that the reform reduced the number of students applying to higher education, and especially discouraged participation of women. After the reform, more men and women applied to STEM programs or non-subsidized non-STEM programs in which they have to pay tuition fees. This latter effect is more larger for women. We proceed by estimating a structural model to analyze how the responsiveness to admission probabilities in application decisions differs between men and women, and find that women are more sensitive to admission probabilities. We estimate the model on a sample before the policy reform and externally validate the model on the cohort affected by the reform. Finally, we use the model to simulate the impact of alternative admission policies on enrollment in STEM programs.
3 May 2018
Giovanna D’Inverno / IMT School for Advanced Studies Lucca/
The effect of additional resources for disadvantaged students: Evidence from a conditional efficiency model /Authors: Kristof De Witte, Mike Smet, Giovanna D’Inverno/
Abstract:
Inequalities in educational and lifelong learning opportunities prevent individuals from reaching their full potential and their personal satisfaction. To reduce the impact of disadvantaged backgrounds on educational achievement, many policies have been promoted. Among them, the Flemish Community of Belgium offers an interesting setting given the inequality level experienced by its educational system. Specifically, the Flemish Ministry of Education enacted a program to provide additional funding to schools with a significant proportion of disadvantaged students. These additional resources are allocated according to an exogenous cut-off. We exploit this information to evaluate the effect of additional funding for disadvantaged students on educational outcomes by using a conditional efficiency analysis. Particular attention is devoted to the impact of socio-economic background variables. Our analysis relies on administrative data on students in secondary education in Flanders.
23 April 2018
Égert Balázs /OECD/
Regulation, institutions and productivity: New macroeconomic evidence from OECD countries
Empirical research on the drivers of multi-factor productivity (MFP) is abundant at the firm- and industry level but surprisingly little research has been conducted on the determinants of MFP at the macroeconomic level. In this paper, we seek to understand the drivers of country-level MFP with a special emphasis on product and labour market policies and the quality of institutions. For a panel of OECD countries, we find that anticompetitive product market regulations are associated with lower MFP levels and that higher innovation intensity and greater openness go in tandem with higher MFP. We also find that the impact of product market regulations on MFP may depend on the level of labour market regulations. Better institutions, a more business friendly environment and lower barriers to trade and investment amplify the positive impact of R&D spending on MFP. Finally, we also show that cross-country MFP variations can be explained to a considerable extent by cross-country variation in labour market regulations, barriers to trade and investment and institutions (including corruption).
19 April 2018
The Tragedy of Complexity
This paper presents an equilibrium theory of product complexity. Complex products generate higher potential surplus, but require more attention from consumers. Because consumer attention is limited and a common resource across products, an attention externality arises: Sellers distort the complexity of their own products to divert attention from other products. In contrast to the classic tragedy of the commons, this externality can lead to too much or too little complexity. Paradoxically, increases in information processing capacity that relax the consumer’s attention constraint can make it more likely that products are overly complex compared with the social optimum.
5 April 2018
Kátay Gábor
Currency Matching and Carry Trade by Non-Financial Corporations
19 March 2018
Lecture:
Darvas Zsolt /Bruegel/: Institutional investors and home bias in financial investments
/Zsolt Darvas and Dirk Schoenmaker/ – February 2018
Abstract
Integrated capital markets facilitate risk sharing across countries. Lower
home bias in financial investments is an indicator of risk sharing.
We highlight that existing indicators of equity home bias in the
literature suffer from incomplete coverage because they consider only
listed equities. We also consider unlisted equites and show that equity
home bias is much higher than previous studies perceived. We also analyse
home bias in debt securities holdings, and euro-area bias. We conclude
that European Union membership may foster financial integration and reduce
information barriers, which sometimes limit cross-country diversification.
We calculate home bias indicators for the aggregate of the euro area as if
the euro area was a single country and report remarkable similarity
between the euro area and the United States in terms of equity home bias,
while there is a higher level of debt home bias in the United States than
in the euro area as a whole.
We develop a new pension fund foreign investment restrictions index to
control for the impact of prudential regulations on the ability of
institutional investors to diversify geographically across borders.
Our panel regression estimates for 25 advanced and emerging countries in
2001-14 provide strong support for the hypothesis that the larger the
assets managed by institutional investors (defined as pension funds,
insurance companies and investment funds), the smaller the home bias and
thereby the greater the scope for risk sharing.
Challenging Science and Innovation Policy Utrecht, 1-3 June 2022, hosted by Copernicus Institute of Sustainable Development, Utrecht University The “European Forum for Studies ... Read More »
Published in ‘Does EU Membership Facilitate Convergence? The Expierience of the EU’s Eastern Enlargement – Volume II’ Edited by Landesmann, Michael, Székely, Istvan P. ... Read More »
Find the full program of the online workshop here Share this:FacebookLinkedInTwitter
