This dissertation presents three essays in financial economics. In the first chapter, Bankruptcy Lawyers and Credit Recovery, I study how bankruptcy law firm advertisements affect credit recovery of households in financial distress. Exploiting the border discontinuity strategy associated with the geographic unit in which local TV advertisements are sold, I empirically uncover bankruptcy filings and credit recovery related to exogenous variations in bankruptcy law firm advertisements. I first document a significant advertising effect on filing rates and show that advertising-induced filers are similar to existing filers. I then find a positive effect of advertisements on credit outcomes including credit score, new homeownership, and foreclosure. I interpret these findings as evidence that lawyers address information frictions in households' assessment of the bankruptcy option. In the second chapter, Cross-subsidization of Bad Credit in a Lending Crisis, we study the corporate-loan pricing decisions of a major, systemic bank during the Greek financial crisis. A unique aspect of our dataset is that we observe both the actual interest rate and the ``breakeven rate'' (BE rate) of each loan, as computed by the bank's own loan-pricing department (in effect, the loan's marginal cost). We document that low-BE-rate (safer) borrowers are charged significant markups, whereas high-BE-rate (riskier) borrowers are charged smaller and even negative markups. We rationalize this de facto cross-subsidization through the lens of a dynamic model featuring depressed collateral values, impaired capital-market access, and limit pricing. In the third chapter, Who Provides Credit in Times of Crisis? Evidence from the Auto Loan Market, we explore lending from traditional banks, credit unions, and finance companies (nonbanks) in the auto loan market over the past two decades with emphasis on the Great Recession and the COVID-19 pandemic. We find that banks provided weak support during the pandemic, thus losing market share and continuing a trend that emerged following the Great Recession. Nonbank market share during this period grew most significantly for subprime borrowers and in counties with stronger bank dependence. We present evidence consistent with a supply-side interpretation of this bank market share loss. These findings contrast with the experience during the Great Recession, when banks contributed the most resilient credit to the auto loan market. Our paper highlights nonbanks’ critical role in the auto loan market in times of crisis, particularly for the subprime segment.

We consider the problem of optimization and decision-making in various settings involving complex systems. In particular, we consider specific problems in decentralized finance which we address employing insights from mathematical finance, in course-mode selection that we solve by applying mixed-integer programming, and in social networks that we approach using tools from graphical game theory.In the first part of the thesis, we model and analyze fixed spread liquidation lending in DeFi as implemented by popular pooled lending protocols such as AAVE, JustLend, and Compound. Empirically, we observe that over 70% of liquidations occur in the absence of any downward price jumps. Then, assuming the borrowers monitor their loans with exponentially distributed horizons, we compute the expected liquidation cost incurred by the borrowers in closed form as a function of the monitoring frequency. We compare this cost against liquidation data obtained from AAVE protocol V2, and observe a match with our model assuming the borrowers monitor their loans five to six times more often than they interact with the pool. Such borrowers must balance the financing cost against the likelihood of liquidation. We compute the optimal health factor in this situation assuming a financing rate for the collateral. Empirically, we observe that borrowers are often more conservative compared to model predictions, though on average, model predictions match with empirical observations. In the second part of the thesis, we consider the problem of hybrid scheduling that was faced by Columbia Business School during the Covid-19 pandemic and describe the system that we implemented to address it. The system allows some students to attend in-person classes with social distancing, while their peers attend online, and schedules vary by day. We consider two variations of this problem: one where students have unique, individualized class enrollments, and one where they are grouped in teams that are enrolled in identical classes. We formulate both problems as mixed-integer programs. In the first setting, students who are scheduled to attend all classes in person on a given day may, at times, be required to attend a particular class on that day online due to social distancing constraints. We count these instances as “excess.” We minimize excess and related objectives, and analyze and solve the relaxed linear program. In the second setting, we schedule the teams so that each team’s in-person attendance is balanced over days of week and spread out over the entire term. Our objective is to maximize interaction between different teams. Our program was used to schedule over 2,500 students in student-level scheduling and about 790 students in team-level scheduling from the Fall 2020 through Summer 2021 terms at Columbia Business School. In the third part of the thesis, we consider a social network, where individuals choose actions which optimize utility which is a function of their neighbors’ actions. We assume that a central authority aiming to maximize social welfare at equilibrium can intervene by paying some cost to shift individual incentives, and that the cost is upper bounded by a budget. The intervention that maximizes the social welfare can be computed using the spectral decomposition of the adjacency matrix of the graph, yet this is infeasible in practice if the adjacency matrix is unknown. We study the question of designing intervention strategies for graphs where the adjacency matrix is unknown and is drawn from some distribution. For several commonly studied random graph models, we show that the competitive ratio of in intervention proportional to the first eigenvector of the expected adjacency matrix, approaches 1 in probability as the graph size increases. We also provide several efficient sampling-based approaches for approximately recovering the first eigenvector when we do not know the distribution. On the whole, our analysis compares three categories of interventions: those which use no data about the network, those which use some data (such as distributional knowledge or queries to the graph), and those which are fully optimal. We evaluate these intervention strategies on synthetic and real-world network data, and our results suggest that analysis of random graph models can be useful for determining when certain heuristics may perform well in practice.

Investments in productive assets by broad-based black economic empowerment (BEE) enterprises in South Africa (SA) during the 1990s have been constrained, in part, by a lack of access to capital. Even if capital can be sourced, BEE businesses often face a liquidity problem, as conventional, equally amortized loan repayment plans do not take into account the size and timing of investment returns, or there are lags in the adjustment of management to such new investments. The aim of this dissertation, therefore, is to compare five alternative loan products to the conventional fixed repayment (equally amortized) loan (FRL) that lenders could offer to finance BEE investments in productive assets that are faced with liquidity stress, namely: the single payment non-amortized loan (SPL); the decreasing payment loan (DP); the partial payment loan (PPL); the graduated payment loan (GPL); and the deferred payment loan (DEFPLO-2). This is done firstly by comparing loan repayment schedules for the six loans using a loan principal of R200 000, repaid over 20 years at a nominal contractual annual interest rate of 10%. Secondly, data from five actual BEE loan applications to ABSA Bank and Ithala in KwaZulu-Natal (KZN) during 2003 are used to compare how the FRL, SPL, DP, GPL, and DEFPLO-l, affect investment profitability, and both the borrower's and the lender's cash-flows, assuming that the lender sources funds from a development finance wholesaler. Results for the first part of the study show that the SPL has smaller initial annual repayments than the FRL (R20 000 versus R23 492) that ease liquidity stress in the early years after asset purchase, but requires a nominal balloon repayment of both interest and principal in year 20 of R220 000. The SPL is also the most costly loan, with total nominal and real repayments that are R130 162 and R43 821, respectively, more than the FRL. The PPL has the lowest total nominal and real repayments assuming that the borrower can make the nominal balloon repayment in year 5 of R202 173. If not, the ending balance of the loan in year 4 would have to be refinanced at current market interest rates. In this situation, the PPL uses very similar financing terms to that of the variable rate long-term loans already used in SA, and thus may not be a useful option to consider for BEE investments facing a liquidity problem. Interest rates may have risen over the last four years of the loan, encouraging lenders to add a premium into the interest rate for the refinanced loan, which could worsen the liquidity position of the BEE enterprise. The DP requires higher initial nominal annual loan repayments (R6 508 more than the FRL) that do not ease the liquidity problem in the early years of operation. The DP loan, however, has total nominal and real repayments that are R59 838 and R23 118, respectively, less than the FRL. A GPL with diminishing, finite interest-rate subsidy seems to have the most potential to ease the BEE investment's liquidity stress. The 17YRGPL used to buy land had total nominal and real repayments that were R84 634 and R67 726 (after subsidy), respectively, less than the FRL. If the GPL was used to purchase machinery-type assets, then the 6YRGPL would have required total nominal and real repayments of R13 957 and R12 596, respectively, less than the FRL. Finally, the DEFPLO-2 loan required a total nominal repayment of R531 128 (R61 290 more than the FRL) and a total real repayment of R345 358 (R26 095 more than the FRL). Clearly, the GPL and DEFPLO-2 loan repayment schedules can partly resolve the liquidity problem in the early years (assuming no major income shocks), although the DEFPLO-2 plan requires higher total repayments than the FRL. The question remains whether lenders would be prepared to implement these two financing plans for BEE investments in productive assets, where the funds to finance the diminishing, finite interest-rate subsidy or the deferment would be sourced, and how the interest-rate subsidy would affect asset values. In the second part of the study, the profitability of the five proposed BEE investments in KZN during 2003 was compared for the five loan products using the Net Present Value (NPV) and the Internal Rateof- return (lRR) capital budgeting procedures. The loan terms, interest rates, principal and characteristics of each BEE firm are different with current rates of return on equity varying by business type. Companies A (five-year loan) and C (10-year loan) are agribusinesses with a higher expected current rate of return of 8% on machinery investments, while companies B (eight-year loan), D (15-year loan), and E (20-year loan) invest in farmland with a lower expected current annual rate of return of 5%. The five business plans may not be representative in a statistical sense of all BEE firms in KZN, but were used because they were readily available. Initially it was assumed that donor/grant funds from a development finance wholesaler were lent to an intermediary (like a commercial bank), which in turn, could finance the five investments using any of the five alternative loans, with the lender's repayment to the wholesaler being via a FRL. It was then assumed that the lender could repay its borrowed funds using the same loans, or combinations of them, that it had granted to these companies. Results show that GPLs and DEFPLs can resolve the liquidity problem associated with investments like land in the early years after purchase provided that projected business performance is adequate, while the SPL and GPL are preferred for BEE projects with stronger initial cash-flows like machinery investments. The study also shows that the loan product that best improves the borrower's liquidity is not always best suited to the lender. In most cases, the GPL suited the borrower, but in four of the five cases, the lender would prefer the SPL and to repay the wholesaler using the SPL. The SPL, however, is unlikely to be used, given the large negative real net cash-flows that it generates when the final payments are due. Recent SA experience with the GPLs (interest rate subsidies funded by private sector sugar millers via Ithala) and the DEFPLs (via the Land Reform Empowerment Facility (LREF) which is a wholesaler of funds in SA) suggests that there is scope to alleviate the liquidity problem if a wholesaler of funds can offer such terms to private banks and venture capital investors who then on-lend to finance BEE asset investments that are otherwise considered relatively high credit risks. This would shift the liquidity problem away from the client to the wholesaler of the funds, but requires access to capital at favourable interest rates. Such capital could be sourced from dedicated empowerment funds earmarked by the private sector, donors and the SA government. The lesson for policymakers is that broad-based BEE could be promoted in other farm and non-farm sectors in SA using similar innovative loan products to complement cash grant funds via financial intermediaries, bearing in mind the limitations of the GPL and DEFPL - such as how to finance the subsidy or deferment, and the impact of income shocks. Donor and National Empowerment Fund capital could be used to allocate grants to provide previously disadvantaged individuals with own equity and also to fund finite, diminishing interest-rate subsidies via GPLs, or to fund DEFPLs (many LREF loans have been leveraged by a cash grant component). This could create an incentive for public/private partnerships, as public/donor funds could be then used to attract private sector funds to finance broadbased BEE investments in SA that satisfy empowerment criteria. The five case studies did not show how the GPLs and DEFPLs could make all profitable (positive net present value) but financially infeasible (returns do not match the size and timing of the lender's financing plan) BEE investments in productive assets under the FRL feasible, except for Company E that showed a positive NPV and IRR when the 19YRGPL was used. They did, however, show how the alternative loans could improve liquidity for investments with either strong or poor cash-flows. The financiers consulted to source case studies in KZN in 2003 at the time of the study could not provide the researcher with any profitable, but financially infeasible, BEE business plans. This raises some concern about how effective these empowerment loan products could be in the future as there is uncertainty over how many potential BEE investments in productive assets in SA are likely to be profitable but financially infeasible. Further research is thus needed to assess the impact of these alternative loans on a wider range of broad-based BEE investments, particularly non-farm projects, than considered in this dissertation. / Thesis (M.Agric.Mgt.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.

The maturity transformation of deposits is a primary driver of economic growth, as loans enable borrowers to spend funds, thereby growing the economy. However, if borrowers cannot repay their loans, the asset quality of banks deteriorate, resulting in non-performing loans or, worse, an economic crisis. An understanding of how macroeconomic and microeconomic determinants impact bank asset quality in South Africa can contribute to knowledge of the bank asset quality phenomenon in the African context. Due to the 2008/2009 global financial crisis, the introduction of new legislation and the value of gold exports, the South African economy presents an opportunity to make an original contribution to the knowledge of determinants that influence bank asset quality. In addition to studying bank asset quality determinants that are contested in research, this study also aims to determine whether a superior returns determinant of non-performing loans exists when comparing a bank’s profitability determinants, namely return on assets, return on equity and interest income on loans. This study applied panel data regression analysis, making use of a balanced panel approach, to study the determinants of bank asset quality. This approach recontextualises the existing bank asset quality theory for the South African financial sector. The results indicate that South Africa is not resilient against the impact of global financial crises trickling through international trade linkages and that regulatory changes do not instantly improve bank asset quality, and may even reduce the short-term asset quality. Moreover, bank asset quality in South Africa is sensitive to the total value of gold exports. It is evident from the profitability measures that the interest income on loans is the most suitable profitability measure of bank asset quality. This study provides an original contribution to bank asset quality determinants and recommends that regulators should pre-emptively determine the impact of new legislation on bank asset quality. Furthermore, interest income on loans as a profitability measure provides the most accurate results. Lastly, a single-country bank asset quality analysis is important, especially for economies that have commodity exports that significantly weigh in on the bank asset mix. / Die termyntransformasie rakende deposito's is die primêre dryfkrag vir groei in die ekonomie: Lenings maak dit vir leners moontlik om fondse te bestee, wat die ekonomie laat groei. Indien hierdie leners hul lenings egter nie kan terugbetaal nie, gaan die gehalte van bankbates agteruit, wat tot wanpresterende lenings of, nog erger, tot 'n ekonomiese krisis kan lei. As begryp kan word hoe makro-ekonomiese en mikro-ekonomiese bepalende faktore op die gehalte van bankbates in Suid-Afrika inwerk, kan dit bydra tot kennis van die verskynsel van bankbategehalte in die Afrika-konteks. In die lig van die 2008/2009 wêreldwye finansiële krisis, die uitvaardiging van nuwe wetgewing en die waarde van gouduitvoere bied die Suid-Afrikaanse ekonomie ’n geleentheid om ’n oorspronklike bydrae te lewer tot kennis van die bepalende faktore wat bankbategehalte beïnvloed. Benewens die bestudering van die bepalende faktore van die gehalte van bankbates wat in navorsing redelik omstrede is, het hierdie studie ten doel om, wanneer 'n bank se winsgewendheidsbepalers, naamlik opbrengs op bates, opbrengs op ekwiteit (eiekapitaal) en rente-inkomste op lenings, met mekaar vergelyk word, vas te stel of daar ’n superieure opbrengsbepaler van wanpresterende lenings bestaan. Vir hierdie studie is ’n regressieontleding van paneeldata uitgevoer, en daar is van ’n gebalanseerde paneelbenadering gebruik gemaak om die bepalende faktore van bankbategehalte te bestudeer. Hierdie benadering herkontekstualiseer die bestaande bankbategehalteteorie vir die Suid-Afrikaanse finansiële sektor. Die resultate van die studie dui daarop dat Suid-Afrika nie veerkragtig is om die uitwerking van wêreldwye finansiële krisisse teen te werk wat met internasionale handelskakelings deursyfer nie en dat reguleringsveranderinge nie dadelik die bankbategehalte verbeter nie; dit kan inteendeel die korttermynbategehalte verlaag. Bowendien is die bankbategehalte in Suid-Afrika gevoelig vir die totale waarde van gouduitvoere. Dit blyk uit die winsgewendheidsmaatstawwe dat die rente-inkomste op lenings die mees geskikte winsgewendheidsmaatstaf van bankbategehalte is. Hierdie studie lewer ’n oorspronklike bydrae tot die bepalers van bankbategehalte en beveel aan dat reguleerders vooruit reeds die uitwerking van nuwe wetgewing op bankbategehalte moet bepaal. Daarby voorsien rente-inkomste op lenings as winsgewendheidsmaatstaf die akkuraatste resultate. Laastens is ’n ontleding van ’n enkele land se bankbategehalte van belang, in die besonder vir ekonomieë met kommoditeitsuitvoere wat beduidend tot die samestelling van bankbates bydra. / Kadimo ya nako ye kopana ya ditipositi ke mokgwa wo bohlokwa wa kgolo ya ekonomi, ka ge dikadimo di dumelela baadimi go šomiša matlotlo, go realo e le go godiša ekonomi. Efela, ge baadimi ba sa kgone go lefela dikadimo tša bona, boleng bja thoto ya dipanka bo a phuhlama, go feleletša go e ba le dikadimo tše di sa šomego gabotse goba, go feta fao, phuhlamo ya ekonomi. Kwešišo ya ka fao ditaetšo tša makroekonomi le maekroekonomi di huetšago boleng bja thoto ya panka ka Afrika Borwa e ka ba le seabe go tsebo ya taba ya boleng bja thoto ya panka go ya ka seemo sa Afrika. Ka lebaka la mathata a ditšhelete a lefase a 2008/2009, tsebišo ya molao wo moswa le boleng bja dithomelontle tša gauta, ekonomi ya Afrika Borwa e fa sebaka seabe sa mathomo tsebong ya ditaetšo tšeo di huetšago boleng bja thoto ya panka. Go tlaleletša nyakišišong ya ditaetšo tša boleng bja thoto ya panka tšeo di ganetšwago nyakišišong, maikemišetšo a nyakišišo ye gape ke go laetša ge eba taetšo ya letseno le legolo la dikadimo tše di sa šomego gabotse di gona ge go bapetšwa ditaetšo tša poelo ya panka, e lego letseno la dithoto, letseno la dišere le letseno la dikadimo. Nyakišišo ye e šomišitše tshekatsheko ya poelomorago ya datha ya phanele, ya go šomiša mokgwa wa phanele wo o lekaneditšwego, go nyakišiša ditaetšo tša boleng bja thoto ya panka. Mokgwa wa go tšwetšapele gape teori ya boleng bja thoto ya panka ya lekala la Afrika Borwa la ditšhelete. Dipoelo di laetša gore Afrika Borwa ga e fokole kgahlanong le khuetšo ya mathata a ditšhelete a lefase ao a rothelago ka dikamanong tša kgwebišano ya boditšhabatšhaba le gore diphetogo tša taolo ga di kaonafatše boleng bja thoto ya panka ka lebelo, gomme di ka fokotša le boleng bja thoto bja paka ye kopana. Go feta fao, boleng bja thoto ya panka ka Afrika Borwa bo ela hloko boleng bja palomoka bja dithomelontle tša gauta. Go a bonagala go tšwa go dikgato tša tiro ya poelo gore letseno la tswala godimo ga dikadimo ke kgato ya poelo ye maleba gagolo ya boleng bja thoto ya panka. Nyakišišo ye e fa seabe sa mathomo ditaetšo tša boleng bja thoto ya panka gomme e šišinya gore balaodi ba swanela go laetša e sa le ka pela khuetšo ya molao wo moswa ka ga boleng bja thoto ya panka. Go feta fao, letseno la tswala godimo ga dikadimo bjalo ka kelo ya tiro ya poelo le go fa dipoelo tše di lebanego gabotse. Sa mafelelo, tshekatsheko ya boleng bja thoto ya panka ya naga e tee, kudu diekonomi tšeo di nago le dithomelontle tša ditšweletšwa tšeo gagolo di dumelelago motswako wa thoto ya panka. / Business Management / Ph. D. (Management Studies)

[pt] Essa tese é composta por 3 artigos. O primeiro estuda a relação entre política monetária e a disponibilidade de crédito direcionado. Com esse propósito, estimamos as respostas de variáveis macroeconômicas setoriais a choques de política monetária identificados com restrições de sinal e restrições de igualdade em um factor-augmented VAR (FAVAR). Nossos resultados mostram que a política monetária perde potência em setores com maior proporção de crédito direcionado. O segundo artigo propõe a introdução de novas restrições para identificação de choques de política monetária. Em particular, além das restrições de sinal usuais sobre taxas de juros e inflação, nós propomos como restrição de identificação que a política monetária não tenha efeitos reais dez anos após o choque. Esse artigo apresenta evidências de que essa restrição é consistente com o modelo novo-keynesiano canônico de 3 equações e com o modelo proposto por Smets and Wouters (2007). Em uma aplicação empírica simples, esse artigo mostra que essa restrição pode ser importante para recuperar efeitos reais de política monetária. O terceiro artigo mostra que bancos estrangeiros podem mitigar barreiras informacionais vis-à-vis bancos privados nacionais a partir da observação do comportamento de seus pares. Dado um pedido de empréstimo por parte de uma firma pequena ou média, a existência de empréstimos bancários passados desta firma com bancos privados nacionais constitui um preditor mais importante da probabilidade de que o empréstimo seja concedido para bancos estrangeiros, em comparação a bancos privados nacionais. Nossos resultados são compatíveis com a visão de que a maior habilidade de bancos privados nacionais em aferir o risco de crédito de firmas pouco transparentes torna os empréstimos anteriores com esses bancos um sinal mais valioso para bancos estrangeiros. / [en] This thesis is composed of three papers. The first one studies the relation between monetary policy power and the availability of earmarked loans. To that end, we estimate the responses of sectoral macroeconomic variables to monetary policy shocks identified through sign and equality restrictions in a factor-augmented VAR (FAVAR). We find that monetary policy loses power in sectors with a larger share of earmarked loans among its bank debt. The second paper proposes the introduction of new restrictions to identify monetary policy shocks in SVARs. In particular, besides standard sign restrictions on interest rates and inflation, we propose to add as an identification restriction the inability of monetary policy to have real effects ten years after the shock. This paper presents evidence of the model consistency of this neutrality restriction both for the canonical 3-equation new keynesian model and the Smets and Wouters (2007) model. In a simple empirical application, this paper shows that this restriction may be important to recover real effects of monetary policy. The third paper shows that foreign banks can mitigate informational barriers vis-a-vis private domestic banks by observing their peers behavior. Conditional on a loan application being filed by a SME firm, we find that the existence of past loans of this firm with private domestic banks constitute a more important predictor that a loan will be granted by foreign banks in comparison to private domestic banks. Our results are compatible with the view that the higher ability of private domestic banks to access informationally opaque business credit risks makes past loans with these banks a more valuable signal for foreign lenders.

No description available.