Estimating Time-Varying Network Effects with Application to Portfolio Allocation

Finance master project by Daniel A. Landau and Gabriel L. Ramos ’19

Editor’s note: This post is part of a series showcasing Barcelona GSE master projects. The project is a required component of all Master’s programs at the Barcelona GSE.

Abstract

In this paper, we characterize a variety of international financial markets as partially correlated networks of stock returns via the implementation of the joint sparse regression estimation techniques of Peng et al. (2009). We explore a number of mean-variance portfolios, with the aim of enhancing out-of-sample portfolio performance by uncovering the hidden network dynamics of optimal portfolio allocation. We find that Markowitz portfolios generally dissuade the inclusion of central stocks in the network, yet the interaction of a stock’s individual and systemic performance is more complex. This motivates us to explore the time-varying correlation of these topological features, which we find are highly market dependent. Building on the work of Peralta & Zareei (2016), we implement a number of investment strategies aimed at simplifying the portfolio selection process by allocating wealth to a targeted subset of stocks, contingent on the time-varying network dynamics. We find that applying mean-variance allocation to a restricted sample of stocks with daily portfolio re-balancing can statistically significantly enhance out-of-sample portfolio performance in comparison to a market benchmark. We also find evidence that such portfolios are more resilient during periods of major macroeconomic instability, with the results applicable to both developed and emerging markets.

Conclusion and Future Research

In our work, we represent 4 international exchanges as individual networks of partially correlated stock returns. To do so, we build a Graph, comprised of a set of Vertices and Edges, via the implementation of the joint sparse regression estimation techniques of Peng et. al (2009). This approach allows us to uncover some of the hidden topological features of a series of Markowitz tangency portfolios. We generally find that investing according to MPT dissuades the inclusion of highly central stocks in an optimally designed portfolio, hence keeping portfolio variances under control. We find that this result is market-dependent and more prevalent for certain countries than for others. From this cross-sectional network analysis, we learn that the interaction between a stock’s individual performance (Sharpe ratio) and systemic performance (eigenvector centrality) can be complex. This motivates us to explore the time-varying correlation ρ between Sharpe ratio and eigencentrality.

Optimal_Weights_for_Tangency_Portfolio_Strategy
Optimal Weights for Tangency Portfolio Strategy.

Overall, we show that in considering the time-varying nature of partially correlated networks, we can enhance out-of-sample performance by simplifying the portfolio selection process and investing in a targeted subset of stocks. We also find that our work proposes a number of future research questions. Although we implement short-sale constraints, it would also be wise to introduce limits on the amount of wealth that can go into purchasing stocks, as this would help to avoid large portfolio variances. Furthermore, our work paves the way for future research into the ability of ρ-dependent investment strategies to enhance portfolio performance in times of macroeconomic distress and major financial crises.

Authors: Daniel A. Landau and Gabriel L. Ramos

About the Barcelona GSE Master’s Program in Finance

Certain uncertainty? The response of Venezuelan banks to a political dilemma

Economics of Public Policy master project by Mary Armijos and Guillem Cuberta ’19

Editor’s note: This post is part of a series showcasing Barcelona GSE master projects. The project is a required component of all Master’s programs at the Barcelona GSE.

Introduction

Our work focuses on the analysis of the Venezuelan banks, which have become more vulnerable. The leading causes of this vulnerability have been oil price shocks, political instability, pro-cyclical monetary conditions (i.e., interest rates), low level of financial intermediation, changing structure (e.g., consolidation, closure or nationalization), non-traditional bank transactions, high exposure to the public sector, and government intervention in their operations.  (Blavy R., 2014) Hence, studying the response of banks to policy uncertainty becomes relevant. However, even more critical, in Venezuela’s context, would be to ask: do banks respond differently to uncertainty if they are politically connected?

To address this question, we construct two indices: the policy uncertainty index and political connection index. Our uncertainty index is in a monthly basis and bank-invariant; it is developed following the work previously done by Baker, Bloom, and David, on the Economic Policy Uncertainty Index (EPU Index) and by Ahir, Bloom, and Furceri, on the World Uncertainty Index (an extension of the EPU).  Similar to Xu and Zhou (2008), for the case of the political connection index, we define the dummy variable connected bank equal one based on whether it is a state-owned bank. Alternatively, in case it is a private one if at least one of the board members has any connection with someone from the government. We adapt these criteria to the information there is available from Venezuela. We also look if one of the members is part of the ‘bolibourgeoisie’ ( a combination of the words Bolivarian and bourgeoisie, a term used to classify the businessmen and public officials linked to the government). For our dependent and micro-control variables (i.e., bank characteristics), we use monthly data from 2006 to 2018 obtained from SUDEBAN (Venezuela’s Superintendence of Banks). Moreover, for the macro-control variables, we obtain monthly data from the International Financial Statistics of the International Monetary Fund (IMF).  The variables we choose for our specifications are based on the works done by Vera et al. (2019) and Bordo et al. (2016).

Our central hypothesis is that when there is high policy uncertainty, political connections may allow connected banks to smooth the effects of uncertainty. We believe that connected banks might respond differently to uncertainty because they have privileged information or preferential treatment from the government, which grants them a competitive advantage over non-connected banks. To test the response of banks, we look at their behavior respect to credit supply and provisions, and we also investigate banks’ profitability in periods of uncertainty through the ROA. Our identification strategy to investigate the causal effect of policy uncertainty and political connection considers the fact that the uncertainty index is a high-frequency time series, which makes it be almost an exogenous variable. Also, the political connection index is not endogenous as it does not vary over time.

In addition to that, we control for bank-invariant and time-specific factors that affect both the right-hand side and left-hand-side variables by adding bank and time fixed effects. Furthermore, to separate the impact of our primary explanatory variable (interaction) from other confounding factors, we control for a block of bank-specific covariates. We also consider different specifications with and without lags of these controls to mitigate the potential reverse-causality concern. Even though we know that all of these adjustments might not entirely correct for omitted variable bias, we consider it adjusts well enough to investigate this relationship. We consider that one of the significant sources of potential bias comes from monthly macro changes (i.e., exogenous shocks like oil prices or U.S. sanctions, and government decisions) that are accounted by including time fixed-effects.

Figure 1: Monthly Economic Policy Uncertainty Index for Venezuela (EPUV)

Results

In our main results, we find that politically connected banks acquire more risks when there is higher uncertainty as an increase of 10 percent in our uncertainty measure leads them to give on average, 0.0262 percent more credits than non-politically connected banks. This result corroborates similar results from the literature that establishes a positive value from being politically connected (Kostovetsky 2015).  Also, we observe that an increase in the uncertainty index induces politically connected banks to hold more loss provisions in their portfolio than non-politically connected banks. A 10 percent increase in our uncertainty index prompts politically connected banks to hold 0.0192 percent more loss provisions than non-politically connected banks. Lastly, the effect of economic policy uncertainty for politically connected banks on ROA has a positive sign. A 10 percent increase in uncertainty increases the average returns on assets of politically connected banks by almost 11 percent compared to non-politically connected banks.

Summing up, connected banks can give more credit to the public in periods of higher uncertainty, at the same time that they hold more loss provisions. The first result is consistent with the ones shown by Cheng et al. (2017), where they find that banks supply much more credit when there is high uncertainty. However, our result of provisions does not coincide with theirs. Contrarily, they find that under higher uncertainty, connected banks reserve lower provisions than unconnected banks. In the case of the profitability analysis, connected banks have lower profits when there is high uncertainty. These results go along with the ones found by Dicko (2016).

We consider that this study presents new relevant findings regarding the literature of political connection and policy uncertainty, and for the Venezuelan economy overall. The political connection matters in periods of high uncertainty but until one point. From our results, we find that politically connected banks seem pro-risk as they give more credit when there is more policy uncertainty. However, on another level, it appears that they do receive privilege information form the government (bad news about the future) that makes them risk-averse at the same time as they also reserve more provisions. Additionally, the result of the relationship between uncertainty and profitability indicators, like ROA, indicates that being politically connected might not be extremely helpful to banks if they only benefit from having more information and do not receive any tangible benefit from the government. For further studies, it would be interesting to analyze how economic agents respond to policy uncertainty depending on the type of benefit they receive from being politically connected to some institutions.

About the Barcelona GSE Master’s Program in Economics of Public Policy

Effective competition in non-workplace pensions

FCA publication with contributions by Lorenzo Migliaccio ’14 (Competition and Market Regulation)

FS19/5 in the context of FCA work across the pension saving value chain . Source: FS19/5

The Financial Conduct Authority has published three pensions papers covering advising on pension transfers, the retirement outcome review, and effective competition in non-workplace pensions. The last one – which I’ve contributed to – outlines a number of proposals to improve competition in the non-workplace pensions market in the UK.

To share my Head of Department’s words, ‘this has been one of the most challenging data gathering exercises I have been involved in’, with more than 100 firms providing input for our analysis.

We found similar weaknesses to those the OFT identified in the DC workplace pension market in 2013, ie demand-side weaknesses and reduced competition on charges.

We now invite stakeholders’ views and welcome alternative suggestions for the way we and the industry can address the issues identified. Here you can find more information and download the feedback statement (pdf).

author

Lorenzo Migliaccio ’14 is Senior Associate Economist at the Financial Conduct Authority. He is an alum of the Barcelona GSE Master’s in Competition and Market Regulation.

LinkedIn | Twitter

Quantification of Instruments’ Strength

Economics master project by Oriol González, Marko Irisarri, Santiago Iglesias, Asier Beristain and Manuel Cabado ’19

Editor’s note: This post is part of a series showcasing Barcelona GSE master projects. The project is a required component of all Master’s programs at the Barcelona GSE.

Abstract

Weak identification is known to yield unreliable standard instrumental variables (IV) inference. A large literature has focused on addressing this issue by proposing methods to detect weak instruments, mainly through the first-stage F-statistic. Our paper evaluates the weak identification in two leading empirical analyses by using the novel alternative approach developed by Ganics, Inoue and Rossi (2018), who base their tests on confidence intervals for the bias of the two-stage least squares estimator, and the size distortion of the associated Wald test. We illustrate the behavior of the tests in empirical settings, and compare how the conclusions differ to those using the standard tests. Our findings suggest that, in our empirical application, the results obtained using this approach are in line with those using previous tests in the literature, confirming it to be a robust alternative. An R package to directly compute these novel tests is also presented.

Main conclusions

The contribution of our paper is mainly twofold. On the one hand, we aim at motivating the usefulness of the novel approach developed by Ganics et al. (2018) to evaluate the robustness of empirical analyses to the potential presence of weak instruments. On the other hand, we make these tests accessible to researchers via the proposed R function. The paper shows how recent tests applied to previous literature unveil new interesting information about the results obtained and hence the conclusions drawn from them. We highlight the consequences of IV estimates displaying both high sampling uncertainty and high specification uncertainty, as minor specification changes can lead to very different estimates, which is in line with current findings in the IV literature (see Yogo, 2004; Kleibergen and Mavroeidis, 2009; Mavroeidis, 2010; Mavroeidis et al., 2014; Ganics, 2017; or Barnichon and Mesters, 2019). Another remarkable issue that arises, mirroring the recent findings in Young (2019), is the importance of the baseline assumptions on the structure of the error variance to correctly interpret the estimation results.

girtest
Output retrieved by the proposed girtest function in R

References

  • Barnichon, R. and Mesters, G. (2019), ‘Identifying modern macro equations with old shocks’, Barcelona GSE Working Paper Series (Working Paper n◦1097).
  • Ganics, G. (2017), Essays in macroeconometrics, PhD thesis, Universitat Pompeu Fabra.
  • Ganics, G., Inoue, A. and Rossi, B. (2018), ‘Confidence intervals for bias and size distortion in IV and local projections-IV models’, Banco de España Working Paper.
  • Kleibergen, F. and Mavroeidis, S. (2009), ‘Weak instrument robust tests in gmm and the new keynesian phillips curve’, Journal of Business & Economic Statistics 27(3), 293-311.
  • Mavroeidis, S. (2010), ‘Monetary policy rules and macroeconomic stability: some new evidence’, American Economic Review 100(1), 491-503.
  • Mavroeidis, S., Plagborg-Møller, M. and Stock, J. H. (2014), ‘Empirical evidence on inflation expectations in the new keynesian phillips curve’, Journal of Economic Literature 52(1), 124-88.
  • Yogo, M. (2004), ‘Estimating the elasticity of intertemporal substitution when instru- ments are weak’, Review of Economics and Statistics 86(3), 797-810.
  • Young, A. (2019), ‘Consistency without inference: Instrumental variables in practical application’, Unpublished manuscript.

About the Barcelona GSE Master’s Program in Economics

Bank Assets, Liquidity and Credit Cycles

Forthcoming publication by Federico Lubello ’12 (Economics)

My paper, “Bank Assets, Liquidity and Credit Cycles” with Ivan Petrella (Warwick and CEPR) and Emiliano Santoro (University of Copenhagen) has been accepted at the Journal of Economic Dynamics and Control. In the paper, we uncover a close connection between the collateralization of bank loans, macroeconomic amplification and the degree of procyclicality of bank leverage.

Abstract

We study how bank collateral assets and their pledgeability affect the amplitude of credit cycles. To this end, we develop a tractable model where bankers intermediate funds between savers and borrowers. If bankers default, savers acquire the right to liquidate bankers’ assets. However, due to the vertically integrated structure of our credit economy, savers anticipate that liquidating financial assets (i.e., loans) is conditional on borrowers being solvent on their debt obligations. This friction limits the collateralization of bankers’ financial assets beyond that of real assets (i.e., capital). In this context, increasing the pledgeability of financial assets eases more credit and reduces the spread between the loan and the deposit rate, thus attenuating capital misallocation as it typically emerges in credit economies à la Kiyotaki and Moore (1997). We uncover a close connection between the collateralization of bank loans, macroeconomic amplification and the degree of procyclicality of bank leverage.

Federico Lubello ’12 is a Research Economist at Banque centrale du Luxembourg. He is an alum of the Barcelona GSE Master’s in Economics.

LinkedIn

Uncertainty in learning, choice and visual fixation

Paper by Hrvoje Stojić (Economics ’11, GPEFM ’17)

source: Stojić et al

Hrvoje Stojić (Economics ’11 and GPEFM ’17) is co-author on a new paper, “Uncertainty in learning, choice and visual fixation,” now available in pre-print on PsyArXiv.

The authors on the paper illustrate the interdisciplinary nature of this research. Hrvoje and co-author Raymond Dolan are researchers at the Max Planck UCL Centre for Computational Psychiatry and Ageing Research; Jacob Orquin of Aarhus University specializes in the role of eye movements in decision making; Peter Dayan is at the Max Planck Institute for Biological Cybernetics), and Maarten Speekenbrink is affiliated with the UCL Department of Experimental Psychology.

About the paper

Hrvoje shares an overview of the paper in this Twitter thread:

Get the pre-print

The paper can be downloaded from PsyArXiv.

alumni

Hrvoje Stojić (Economics ’11, GPEFM ’17) is a researcher at UCL. He is an alum of the Barcelona GSE Master’s in Economics and PhD from GPEFM (UPF and Barcelona GSE).

LinkedIn | Twitter | Github

City Design, Planning, Policy Innovations: The Case of Hermosillo

IDB publication co-authored by Miguel Angel Santos (ITFD ’11, Economics ’12)

credit: IDB

After a lengthy review process we are proud to announce that our book “City Design, Planning, Policy Innovations: The Case of Hermosillo” is published and available for download from the Inter-American Development Bank. Cutting edge research on cities featuring my work with Douglas Barrios, my colleague at the Center for International Development’s Growth Lab at the Harvard Kennedy School. Thanks to Andreina Seijas and Diego Arcia for the superb coordination and editing work.

About the book

This publication summarizes the outcomes and lessons learned from the Fall 2017 course titled “Emergent Urbanism: Planning and Design Visions for the City of Hermosillo, Mexico” (ADV-9146). Taught by professors Diane Davis and Felipe Vera, this course asked a group of 12 students to design a set of projects that could lay the groundwork for a sustainable future for the city of Hermosillo—an emerging city located in northwest Mexico and the capital of the state of Sonora. Part of a larger initiative funded by the Inter-American Development Bank and the North-American Development Bank in partnership with Harvard University, ideas developed for this class were the product of collaboration between faculty and students at the Graduate School of Design, the Kennedy School’s Center for International Development and the T.H. Chan School of Public Health.

Miguel Angel Santos (ITFD ’11, Economics ’12) is Director of Applied Research at the Growth Lab at Harvard Kennedy School.

LinkedIn | Twitter

The UK productivity puzzle

Speech by Gavin Jackson ’12 (Economics) to the Oxford Economics Society

Image: OES

This June, Gavin Jackson ’12 (Economics) returned to his undergrad alma mater, University of Oxford, and gave a talk to the Oxford Economics Society about the slowdown in productivity in the United Kingdom and where productivity in the UK might be headed.

He listed five contributing factors to the slowdown: “changes in financial regulation, the patent cliff, mismeasurement of telecommunications, attempts to cope with climate change, and the troubles with getting more oil out of the North Sea.”

Looking ahead, he remarked, “I don’t think we can or should go back to the past. We do not want to go back on environmental on financial regulation, as the US is doing right now. But what we can do as a society is try to be open to new opportunities and technologies that are coming along and that means investing in the basics of education, infrastructure and research to make sure that we are able to make the most of things like e-commerce and working out what to do about those who lose out from these transitions.”

Gavin Jackson ’12 is an Economics Reporter at the Financial Times. He is an alum of the Barcelona GSE Master’s in Economics.

LinkedIn | FT Articles

The Zero Lower Bound was irrelevant

Blog post for AIER by Brian C. Albrecht ’14 (Economics of Public Policy)

empty building floor

Brian Albrecht is a PhD candidate at the University of Minnesota and a graduate of the Barcelona GSE Master’s Program in Economics of Public Policy, as well as a past editor of the Barcelona GSE Voice. He is also a contributor to the Sound Money Project, a blog from the American Institute for Economic Research (AIER).

In a recent post, Brian talks about a recent paper by Barcelona GSE professors Davide Debortoli, Jordi Galí, and Luca Gambetti, “On the Empirical (Ir)Relevance of the Zero Lower Bound Constraint.” He writes:

Many economics writers, including Ben BernankeNeil Irwin, and Justin Wolfers, worry that the Fed will not be able to combat the next recession. Current interest rates, the sad story goes, are already close to zero. Since a downturn will push the economy to the zero lower bound (ZLB), the Fed will not be able to lower rates further, thereby prolonging the recession.

Of course, for such a story to make sense, the ZLB must be a fundamental constraint that inhibits monetary policy. In a new NBER working paper, Davide Debortoli, Jordi Galí, and Luca Gambetti consider whether the ZLB was actually the problem during the last recession. They say the ZLB was irrelevant. The authors come to this conclusion by studying two types of evidence: measures of macro volatility, and the response of macro variables to aggregate shocks through a vector autoregression.

Brian C. Albrecht for Sound Money

Read Brian’s full post on this paper and find a list of all his recent posts over on the AIER website.

alumni

Brian C. Albrecht ’14 is a PhD candidate in Economics at the University of Minnesota. He is an alum of the Barcelona GSE Master’s in Economics of Public Policy.

Twitter | Website

Women in Economics seminar

Master’s students Analía García ’19 (ITFD) and Lorena Franco ’19 (Economics) organized the seminar to highlight research by female PhD students and professors

Women in Economics seminar

This May, BGSE Master’s students Analía García ’19 (ITFD) and Lorena Franco ’19 (Economics) organized the Women in Economics two-day seminar, which meant to highlight female PhD students and faculty members’ research.

Three students and four Barcelona GSE Affiliated Professors presented their work, which varied from family economics to political economics and experimental economics. More information of the speakers and their topics below.

Organizers Analía García ’19 and Lorena Franco ’19

These efforts, nonetheless, started over two months ago when both students, who are from Latin America and the Caribbean, organized an open forum on International Women’s Day. Having prior work experience and noting the clear lack of female representation in economics and academia, they wanted to expand the conversations on the topic and discuss what we could do to potentially “make it better” within their parameters. The Women in Economics seminar was born from the conversations during the first and second open forums, and thanks to the ideas of Marta Morazzoni and Claudia Meza, both PhD students at GPEFM (UPF and Barcelona GSE).

Putting this together was a challenge given this had not been done at BGSE before, but the organizers hope this was insightful for all those who attended.

More female and racial diversity in economics and academia, please!

The speakers and the titles of the work were the following (listed alphabetically):

PhD Students

  • Marta Morazzoni “Family Dynamics in Macroeconomics: when the representative household does not represent us anymore”
  • Marta Santamaría “The Gains from Reshaping Infrastructure: Evidence from the Division of Germany”
  • Alina Velias “When to Tie Odysseus to the Mast: Costly Commitment Under Biased Expactations”

Professors

  • Enriqueta Aragonés “Stability of a Multi-level Government: A Catalonia in Spain”
  • Rosa Ferrer “Consumers’ Costly Responses to Product-Harm Crises” and “Gender Gaps in Performance: Evidence from Young Lawyers”
  • Ada Ferrer-i-Carbonell “Relative Deprivation in Tanzania”
  • Rosemarie Nagel “Regularities in the Lab, Brain, and Field: A Cognitive Reasoning Model”