Working Papers
Forecast combination in the frequency domainAuthor(s): Gonçalo Faria, Fabio Verona Date: January 2024 Abstract: We propose a new forecasting method – forecast combination in the frequency domain – that takes into account the fact that predictability is time and frequency dependent. We use this method to forecast the equity premium and real GDP growth rate. Combining forecasts in the frequency domain produces markedly more accurate predictions relative to the standard forecast combination in the time domain, both in terms of statistical and economic measures of out-of-sample predictability. In a real-time forecasting exercise, the flexibility of this method allows to capture remarkably well the sudden and abrupt drops associated with recessions and further improve predictability. |
||||
Courts performance: The case of PortugalAuthor(s): Nuno Q. M. Lopes, Maria C. A. Silva Date: December 2023 Abstract: For some time now, the judicial systems of countries have been intrinsically connected to their development. This analysis uses data from 2015 to 2021 on every bench in the Portuguese first instance courts, consisting of 3,249 observations and 12 different variables. To analyze the efficiency of benches, Data Envelopment Analysis was used and scale and scope economies were investigated. The results showed that the efficiency of benches was very balanced throughout the country and over the years. Given the clear difference between specialized and non-specialized benches, we assessed benches within and between groups and computed the group frontier gap. This allowed us to draw conclusions regarding the economies of scope of specialized vs generic benches. The results uncovered diseconomies of scope (generic benches are less efficient than specialised ones), and diseconomies of scale with some specialised benches experiencing decreasing returns to scale (criminal, enforcement and commerce benches) while others experienced increasing returns (Civil and generic benches). |
||||
The pricing of sustainable syndicated loansAuthor(s): Paulo Alves, Jorge Gonçalo, João Pinto Date: November 2023 Abstract: This paper provides a comparative analysis of sustainable and conventional syndicated loan spreads and pricing. Using a cross-section of 24,962 syndicated loan tranches closed between 2018 and 2022 in OECD countries, we show that sustainable and conventional loans are differently priced, spreads of sustainable versus conventional loans do not differ significantly, and banks rely on contractual, macroeconomic, bank syndicate structure, and borrowers’ characteristics when pricing sustainable tranches. At the deal-level, our results do not support the hypothesis of sustainable debt financing as a mechanism for reducing firms’ funding costs. We also find that economies of scale, institutional, and information asymmetry arguments affect firms’ choice between sustainable and conventional syndicated deals. |
||||
Common-Ownership vs. CrossOwnership: Evidence from the Automobile IndustryAuthor(s): Cristian Huse, Ricardo Ribeiro, Frank Verboven Date: October 2023 Abstract: Overlapping ownership has gained considerable momentum in the last decades, yet little is known about the role of its sources. We quantify the relative importance of common-ownership, by shareholders external to an industry, and cross-ownership, by firms within the industry. We focus on the global automobile industry, over the period 2007-2021, and document that common-ownership links amount to 31–39%, while crossownership links amount to 5–9% of automobile manufacturers’ stock. We show that not accounting for these relatively modest cross-ownership links has important implications: it can increase the average weight assigned by managers to the profit of competitors by between 33–68%. |
||||
Adult-Use Cannabis Regulation: Contribution to a Scientifically Informed PolicyAuthor(s): Ana Lourenço, Maria Carmo Carvalho, Teresa Summavielle Date: February 2023 Abstract: On 15th July 2022, the Ministers (or high representatives) in the fields of drug supply and demand of the governments of Germany, Luxembourg and Malta issued a joint statement on the regulation of cannabis for non-medical and non-scientific uses. The document states that “there is a need to re-assess our policies on cannabis and to take into account recent developments in this area, to further strengthen and develop health and social responses, such as prevention programs, treatment and harm reduction interventions and to find new approaches beyond prohibition-based drug policies”. This statement is a2 step forward, in Europe, towards the adoption of adult-use cannabis regulation as an alternative policy vis-a-vis prohibition. |
||||
Innovation through inter-regional interaction in a spatial economic modelAuthor(s): José M. Gaspar & Minoru Osawa Date: November 2022 Abstract: This paper analyses a two-region model with vertical innovations that enhance the quality of varieties of the horizontally differentiated manufactures produced in each of the two regions. We look at how the creation and diffusion of knowledge and increasing returns in manufacturing interact to shape the spatial economy. Innovations occur with a probability that depends on the inter-regional interaction between researchers (mobile workers). We find that, if the weight of interaction with foreign scientists is relatively more important for the success of innovation, the model accounts for re-dispersion of economic activities after an initial stage of progressive agglomeration as transport costs decrease from a high level. |
||||
The price impacts of the exit of the Hazelwood coal power plantAuthor(s): Ricardo Gonçalves & Flávio Menezes Date: October 2022 Abstract: This paper estimates the price impacts of the unanticipated closure of Hazelwood, a large brown coal power plant (1600 MW) in Victoria, Australia. We measure the total impact of the closure on prices in Australia's National Electricity Market for each half-hour interval and for each state 3 months, 6 months, and 12 months from closure. We also break down the impact into direct and indirect efects. We find that the total impact of the closure on prices varies considerably across half-hours. The results vary not only in magnitude and across time, but also in statistical significance. Our estimates suggest an upper bound for the impact on the average half-hourly price of $18.90/MW 12 months from closure, with a total market impact of $4,287.7 million. When we break down the total impact into direct and indirect e¤ects, we find the latter to be the main driver of our results. In particular, we find that the reduction in the prices because of increased wind generation in a given half hour - the merit-order efect - has decreased markedly following the closure, and this largely explains the observed price increases post-closure. |
||||
Judging a wine by its bottle: The influence of wine packaging on consumers’ responsesAuthor(s): Maria Teresa Brito e Faro & Carla Carvalho Martins & Joana César Machado Date: September 2022 Abstract: Packaging plays a key role in a product’s success, affecting consumers’ attitudes and purchase intentions. For wine, the role of packaging is even more relevant. Indeed, wine packaging is critical for making wine attributes and perceived quality tangible through visual elements, such as the bottle and label shapes. The present research aims to understand how bottle and label shapes influence consumers' quality perceptions, affect towards wine packaging and, consequently, consumers' purchase intention. |
||||
Scalable Probability Vector MachinesAuthor(s): Pedro Duarte Silva Date: July 2022 Abstract: Over the last few decades, kernel based methods have added an important set of tools to any statistician or econometrician, providing competitive, model free, alternatives to traditional methodologies. In particular, classification Support Vector Machines (SVMs) have proven to be among the most accurate predictors of unknown class memberships in supervised classification problems. Unfortunately, standard SVMs do not complement these predictions with reliable estimates of the corresponding class membership probabilities. Recently, it has been shown that sequences of nonstandard (weighted) SVMs can overcome this limitation, and may be used to recover precise class probability estimates. However, existing implementations of this approach can be computationally too demanding, and may not scale well when the number of diferent classes grows. In this work, we will present an improved method to build k-class probability estimates from sequences of weighted SVMs, with good scaling properties as k increases. Simulation experiments show that class probability estimation based on weighted SVMs is often more accurate than competing distribution free machine learning approaches, and are more reliable than multinomial logisic regression when its assumptions fail. A public domain R package implementing this proposal is under preparation. |
||||
Estimating Oligopoly with Shareholder Voting ModelsAuthor(s): José Azar & Ricardo Ribeiro Date: June 2022 Abstract: We develop an empirical model of overlapping ownership conduct. The model (i) links firm conduct parameters to deep parameters of the firm’s process of shareholder preference aggregation through voting; (ii) can cope with ownership settings involving both intra- and inter-industry overlapping ownership; and (iii) yields an equilibrium flexible formulation for the management’s objective function that allows for no internalization, partial internalization and full internalization of shareholder objectives by managers. Using data for the U.S. airline industry in the 2015-2017 period, we find evidence for a partial internalization formulation in which managers put significant weight on shareholder objectives, but substantially less than in the full-internalization limiting case. We find also that inter-industry overlapping ownership is associated to lower inferred marginal costs, and that omitting inter-industry overlapping ownership leads to substantial bias towards zero in the parameters that drive how much intraindustry overlapping ownership is internalized by the firms. Finally, we find, focusing on the 2017Q4 period, that overlapping ownership overall (both intra- and inter-industry) seems to increase the average airline fare by 4.0%, increase industry profit by 24.4% and decrease consumer surplus by 1.8%, and that these effects are mostly due to overlapping ownership by shareholders other than the “Big Three” asset managers. |
||||
Project Finance Bonds: An Empirical Analysis of Spread and Choice DeterminantsAuthor(s): Sebastião S. Guedes & João M. Pinto Date: May 2022 Abstract: This paper provides a comparative analysis of project finance (PF) and traditional corporate finance (CF) bond pricing, along with an examination of the choice determinants of PF vis-à-vis CF transactions. Using a crosssection of 47,196 bonds issued worldwide in the 1993-2020 period, we show that PF and CF bonds are differently priced, PF bonds have higher spreads than comparable CF bonds, and although ratings are the most important pricing determinant for PF and CF bonds at issuance, investors rely on other contractual, macroeconomic, and firms’ characteristics beyond these ratings. We find that economies of scale, risk management, and information asymmetry arguments affect sponsoring firms’ choice between PF and CF transactions. Findings suggest that firms choose PF when they are relatively smaller, less profitable and creditworthy, and seek long-term inancing. Our results also document that the host country’s legal and institutional frameworks affect financing choices. |
||||
The Choice between Project Financing and Corporate Financing: Evidence from the Corporate Syndicated Loan MarketAuthor(s): Paulo P. Alves & João M. Pinto Date: May 2022 Abstract: We examine the factors that influence public firms’ choice between project finance and internally organized investment projects. Using a large sample of syndicated deals closed between 2000 and 2020 in conjunction with Datastream data, we find that economies of scale, agency costs of debt, and information asymmetry arguments affect the choice of on- versus off-balance-sheet funding. As project finance deals have higher borrowing costs than comparable corporate financing deals, we show that other firm-level countervailing benefits play a key role in the sponsoring firms’ choice: firms choose project over corporate financing when they are relatively larger, less profitable and creditworthy, and seek long-term financing; and switchers - firms that closed the two types of deals during the sample period - that resort to project finance tend to be more levered and have larger growth opportunity sets. |