Title: Investments And Innovation With Non-Rival Inputs: Evidence From Chinese Artificial Intelligence Startups

Fields: Innovation And Entrepreneurship

Supervisors: Daniel Sgroi, James Fenske

Title: Worldwide Child Stunting since the Nineteenth Century

Abstract: This paper conducts a meta-analysis of 877 historical child growth studies to reconstruct child stunting rates, the share of children who are too short for their age, for 121 countries from the earliest date possible to the present. This data complements and extends the modern Joint Malnutrition Estimates database of country-level stunting rates, which begins in the 1980s. We find that many European countries had stunting rates similar to current LMICs at the turn of the twenti- eth century, but child stunting fell in the early twentieth century reaching very low levels before World War II. Stunting rates were also very high in Japan and Korea. However, stunting rates were surprisingly low historically in the European settler colonies, Eastern Europe and the Caribbean. Historical comparisons of child stunting add a new dimension to the historical health transition and allow for more direct historical lessons for the fight against stunting today.

Title: When Nation Building Goes South: Draft Evasion, Government Repression, And The Origins Of The Sicilian Mafia

Fields: Economic History, Political Economy

Supervisors: James Fenske, Sharun Mukand

Title: Solving the Forecasting Combination Puzzle (with Ryan Zischke, Gael M. Martin and Donald Poskitt)

Abstract: We demonstrate that the so-called forecasting combination puzzle is a consequence of the methodology commonly used to produce forecast combinations. By the combination puzzle, we refer to the empirical finding that predictions formed by combining multiple forecasts in ways that seek to optimize forecast performance often do not out-perform more naive, e.g. equally-weighted, approaches. In particular, we demonstrate that, due to the manner in which such forecasts are typically produced, tests that aim to discriminate between the predictive accuracy of such competing combinations can have low power, and can lack size control, leading to an outcome that favors the simpler approach. In short, we show that this counter-intuitive result can be completely avoided by the adoption of more efficient estimation strategies in the production of the combinations, when feasible. We illustrate these findings both in the context of forecasting a functional of interest and in terms of predictive densities.

Title: School Capital Expenditure Rules, Student Outcomes, and Real Estate Capitalization (joint work with Barbara Biasi and Julien Lafortune)

Abstract: School capital expenditures are a major component of public spending in the US, but there is substantial disagreement among economists and policymakers under what conditions these expenditures are a good use of funds. We use new data on school bond elections from dozens of states to estimate (a) the effects of capital expenditures on student achievement and house prices, (b) how they vary with bond and district characteristics, and (c) how they are shaped by the institutional environment. On average, we find positive effects on test scores and house prices, but with meaningful heterogeneity across bond characteristics, districts, and state institutions. We interpret these findings in a probabilistic voter model with budget-maximizing school districts. The model provides a foundation for quantitative evaluation of counterfactual political and fiscal institutions governing school capital finance.

Title: When Nation Building Goes South: Draft Evasion, Government Repression, And The Origins Of The Sicilian Mafia

Fields: Economic History, Political Economy

Supervisors: James Fenske, Sharun Mukand

Title: Adaptable Economics Education - What, Why and How (joint with Cloda Jenkins, Imperial College)

Abstract: In this presentation, we provide guidance for instructors looking to move to an adaptable learning design for their economics courses, as well as preliminary evidence of its evaluation at UCL. We use the term “adaptable” as it embraces both online and face-to-face elements and is designed to make moving in between the two delivery methods relatively easy. This is of course particularly relevant in light of the pandemic disruption but may also be considered to be a helpful direction of travel in an uncertain world in order to make the higher education system more resilient to shocks. In this context, “adaptable” is not equivalent to hybrid, that is, teaching online and in-person students simultaneously. The two main reasons for this are that we aim to include multiple models of adaptable education here, which could include a hybrid option, and that in general hybrid models are the most difficult to deliver successfully and should be avoided where possible. 

Organised by Subhasish Dey

Title: Dynamic Contracting with Multiple Agents under Limited Commitment

We consider an environment of dynamic contracting with multiple agents and lack of commitment. A principal with no commitment power would like to screen efficient workers over time and assign harder tasks to them. After efficiency is revealed, the principal becomes tempted to change the terms of trade. . Breaches of contracts are observable and, hence, whenever past promises are not honored future information revelation stops. We provide necessary and sufficient conditions under which the principal is able to foster information revelation. Optimal contracts that lead to information revelation entail extreme high-powered incentives after information is revealed, and rewards for information revelation disappear in the long run. Information revelation becomes easier when workers are stochastically replaced by new ones.

Title: Worker Sorting And The Gender Wage Gap

Fields: Labour

Supervisors: Roland Rathelot, Manuel Bagues

Title: Belief-Driven Fluctuations, Global Imbalances And Equilibrium Dynamics

Fields: Macro, Theory

Supervisors: Roger Farmer, Herakles Polemarchakis, Pablo Beker

Title: The Revolution Suffocates Its Children - The Short- and Long-Term Effects of Air Pollution in Socialist East Germany

Abstract: Measuring the detrimental effects of air pollution on individuals is difficult. In this paper, we overcome this challenge by leveraging a natural experiment occurring in socialist East Germany in the 1980s. Suddenly and unexpectedly, the Soviet Union reduced and capped East Germany’s access to imported fossil fuels, leading the socialist party dictatorship to rapidly substitute Soviet oil with domestic brown coal at the cost of increased ambient air pollution. Comparing regions within East Germany with and without natural brown coal deposits, we find that the switch to brown coal led to an immediate and permanent increase in mortality (?), infant mortality and a reduction in birth weights. We use administrative social security data after German reunification to show that, in the next 40 years, individuals that lived in areas within the GDR exposed to the shift to brown coal spent less time in employment, earned lower wages and retired earlier. The authoritarian nature of the East German government makes this natural experiment particularly insightful by ruling out spatial sorting behaviour, competitive housing markets, and labour market adjustments as channels through which the estimated effects of air pollution could have been confounded.

Title: Can An Ai Agent Hit A Moving Target

Fields: Macro, Theory

Supervisors: Roger Farmer, Herakles Polemarchakis

Title: Heterogeneous Local Projections: An Application To Firm Investment

Fields: Macro, Econometrics

Supervisors: Giovanni Ricco, Ana Galvao (Wbs)

Title: Predicting Long-term Unemployment Risk - Andreas Mueller (UT Austin) & Johannes Spinnewijn (LSE)

Abstract: This paper uses rich administrative and survey data from Sweden to study the predictability and determinants of long-term unemployment (LTU) over the period 1992-2016. We use standard machine learning techniques to predict job seekers' LTU risk and find substantial predictable heterogeneity. Compared to a model using standard socio-demographic variables, a comprehensive model that uses data on income, employment and benefit histories more than doubles the predictive power. The estimated heterogeneity in LTU risk implies that at least two thirds of the observed duration dependence in job finding is driven by dynamic selection. We apply our prediction algorithm over the business cycle and find significant heterogeneity underlying the cyclicality in average LTU risk, while the role of composition effects is limited. We evaluate the implied value of targeting unemployment policies and how this changes over the business cycle.

Title: The Fertility Impact Of Mass Return Migration

Fields: Economic History, Labour

Supervisors: James Fenske, Manuel Bagues

Title: How informed do you want your principal to be? (joint with Rahul Deb and Mallesh Pay)

Title: Optimal Political Career Dynamics

Title: Avoiding Dissonant Information

Fields: Behavioural Economics, Experimental Economics

Supervisors:

Title: Price Changes And Welfare Analysis: Measurement Under Individual Heterogeniety

Fields: Theory, Public, Labour

Supervisors: Herakles Polemarchakis, Robert Akerlof, Costas Cavounidis

Title: Eigenvalue tests for the number of latent factorsin short panels

Alain-Philippe Fortin, Patrick Gagliardini, Olivier Scaillet
This paper studies new tests for the number of latent factors in a large cross-sectionalfactor model with small time dimension. These tests are based on the eigenvalues ofvariance-covariance matrices. We establish the asymptotic distributional results usingexpansion theorems based on perturbation theory for symmetric matrices. Our frame-work accommodates semi-strong factors in the systematic components. We propose anovel statistical test for weak factors against strong or semi-strong factors. We providean empirical application to US equity data. Evidence for a dierent number of latentfactors according to market downturns and market upturns, is statistically ambiguous.In particular, our results contradicts the common wisdom of a single factor model inbear markets

Title: The Education-Innovation Gap.

Title: Using an Online Interactive Textbook for Content Delivery in a large on-campus Mathematics for Economists course.

The Covid pandemic forced lecturers to rethink the way content is delivered to students. Live online lectures and pre-recorded video recordings became standard tools of synchronous and asynchronous content delivery respectively. Reading materials (textbooks or lecture provided notes) also often supplement the content delivery, but are not usually used as the sole source of new content.

We present summary findings regarding the use of an online, interactive textbook-type resource. This resource combined text-based content delivery with interspersed pre-recorded online videos and questions (with instantaneous feedback to students) to check students’ understanding. This resource was produced for a large Mathematics for Economists unit with 900+ students in the pandemic affected academic years 2020/21 and 2021/22. In the second of these years this resource was used despite on-campus lectures being available again as a tool.

While the online textbook was used to deliver the content asynchronously, there were twice-weekly Review and Q&A sessions in which students could bring up any questions and problems and in which the lecturer would review some of the key concepts introduced in the online lessons.

We present findings on the usage pattern of the resource as well as student feedback. The student feedback presented, leads to the conclusion that, even as on-campus lecture delivery is becoming available again, a future blend of learning activities can include substantial asynchronous content delivery. This flipped classroom setup is attractive as it can create the space in synchronous meetings to focus on active student learning which has proved to contribute positively to learning outcomes (Freeman et al . (2014).

The online textbook allows us to observe patterns of study we cannot normally observe. The amount of time spend on the learning resource and its timing as well as the engagement with the feedback questions allows us to identify some of the study patterns of successful and less successful students.

Organiser: Subhasish Dey

Panel:
Juliette de Wit (University of Groningen)
Melanie Fox (Purdue University)
Miguel Flores (National College of Ireland)

Chair: Cloda Jenkins (Imperial)

Co-organiser: Stefania Paredes Fuentes

to attend on Zoom, register here before Monday 24th October.
The session will be live-streamed on the CTaLE YouTube Channel.

TItle: Brokering Votes with Information Spread via Social Networks (with R Duarte, F Finan, H Larreguy)

Title: Bounds on Average Effects in Discrete Choice Panel Data (joint with Cavit Pakel).

Abstract: Average effects in discrete choice panel data models with individual-specific fixed effects are generally only partially identified in short panels. While consistent estimation of the identified set is possible, it generally requires very large sample sizes, especially when the number of support points of the observed covariates is large, such as when the covariates are continuous. In this paper, we propose estimating outer bounds on the identified set of average effects. Our bounds are easy to construct, converge at the parametric rate, and are computationally simple to obtain even in moderately large samples, independent of whether the covariates are discrete or continuous. We also provide asymptotically valid confidence intervals on the identified set. Simulation studies confirm that our approach works well and is informative in finite samples. We also consider an application to labor force participation.

Title: Age Set vs. Kin: Culture and Financial Ties in East Africa.

Abstract: We study how social organization shapes patterns of economic interaction and the effects of national policy, focusing on the distinction between age-based and kin-based groups in sub-Saharan Africa. Motivated by ethnographic accounts suggesting that this distinction affects redistribution, we ana- lyze a cash transfer program in Kenya and find that in age-based societies there are consumption spillovers within the age cohort, but not the extended family, while in kin-based societies we find the opposite. Next, we document that social structure shapes the impact of policy by showing that Uganda’s pension program had positive effects on child nutrition only in kin-based societies.

Title. Weighted representative democracy

Abstract. I will discuss a new system of democratic representation. In this system, any voter can choose any legislator as her representative; thus, different legislators can represent different numbers of voters. Decisions in the legislature are made by weighted majority voting, where the weight of each legislator is determined by the number of voters she represents. In the first part of the talk, I will discuss some joint work with Arnold Soh, in which we show that, if the size of the electorate is very large, then with very high probability, the decisions obtained in the legislature will agree with those which would have been reached by a popular referendum decided by simple majority vote. In the second part of the talk, I will discuss my more recent work on endogenous candidacy in this model (this is work in progress).