Search

The secured borrowing based on sell/buy-backs agreements is studied, specifically considering both: quantity and price. The empirical evidence presented in this paper suggests that, after controlling for specific individual characteristics, group-specific effects (defined by belonging or not to a financial group) play a relevant role in this market. Using spatial panel data models, we find that the amount of liquidity obtained with sell/buy-backs depend on traditional determinants (institution’s size and financial leverage), but also, on the average size of the financial group to which the financial institution belongs. Similarly, the borrowing cost depends on the amount of liquidity, but the average profitability of the financial group is also significant. Our results are robust to different relationship structures specified for financial groups.

This paper employs network analysis to study world trade from 1995 to 2014. We focus on the main connective features of the world trade network (WTN) and their dynamics. Results suggest that countries’ efforts to attain the benefits of trade have resulted in an intertwined network that is increasingly dense, reciprocal, and clustered. Trade linkages are distributed homogeneously among countries, but their intensity (i.e. their value) is highly concentrated in a small set of countries. The main connective features of the WTN were not affected by the 2007-2008 international financial crisis. However, we find that the crisis marks a turning point in the evolution of the WTN from a two-group (led by the US and Germany) to a three-group (led by the US, Germany, and China) hierarchical structure; gravity models of international trade may explain this evolution. Furthermore, we find that WTN’s connective features do not conform to a linear aggregation of sectorial trade networks.

The balance sheet is a snapshot that portraits the financial position of a firm at a specific point of time. Under the reasonable assumption that the financial position of a firm is unique and representative, we use a basic artificial neural network pattern recognition method on Colombian banks’ 2000-2014 monthly 25-account balance sheet data to test whether it is possible to classify them with fair accuracy. Results demonstrate that the chosen method is able to classify out-of-sample banks by learning the main features of their balance sheets, and with great accuracy. Results confirm that balance sheets are unique and representative for each bank, and that an artificial neural network is capable of recognizing a bank by its financial accounts. Further developments fostered by our findings may contribute to enhancing financial authorities’ supervision and oversight duties, especially in designing early-warning systems.

We examine how liquidity is exchanged in different types of Colombian money market networks (i.e. secured, unsecured, and central bank’s repo networks). Our examination first measures and analyzes the centralization of money market networks. Afterwards, based on a simple network optimization problem between financial institutions’ mutual distances and number of connections, we examine the tradeoff between liquidity risk and counterparty risk. Empirical evidence suggests that different types of money market networks diverge in their centralization, and in how they balance counterparty risk and liquidity risk. We confirm an inverse and significant relation between counterparty risk and liquidity risk, which differs across markets in an intuitive manner. We find evidence of liquidity cross-underinsurance in secured and unsecured money markets, but they differ in their nature. Central bank’s role in mitigating liquidity risk is also supported by our results.

The effect of the Global Financial Crisis (GFC) has been substantial across markets and countries worldwide. We examine how the GFC has changed the way equity markets group together based on the similarity of stock indices’ daily returns. Our examination is based on agglomerative clustering methods, which yield a hierarchical structure that represents how stock markets relate to each other based on their cross-section similarity. Main results show that both hierarchical structures, before and after the GFC, are readily interpretable, and indicate that geographical factors dominate the hierarchy. The main features of equity markets’ hierarchical structure agree with most stylized facts reported in related literature. The most noticeable change after the GFC is a stronger geographical clustering. Some changes in the hierarchy that do not conform to geographical clustering are explained by well-known idiosyncratic features or shocks.

An interacting network coupling financial institutions' multiplex (i.e. multi-layer) and financial market infrastructures' single-layer networks gives an accurate picture of a financial system’s true connective architecture. We examine and compare the main properties of Colombian multiplex and interacting financial networks. Coupling financial institutions' multiplex networks with financial market infrastructures' networks removes modularity, which enhances financial instability because the network then fails to isolate feedbacks and limit cascades while it retains its robust-yet-fragile features. Moreover, our analysis highlights the relevance of infrastructure-related systemic risk, corresponding to the effects caused by the improper functioning of FMIs or by FMIs acting as conduits for contagion.

We study the network of Colombian sovereign securities settlements. With data from the settlement market infrastructure we study financial institutions’ transactions from three different trading and registering individual networks that we combine into a multi-layer network. Examining this network of networks enables us to confirm that (i) studying isolated single-layer trading and registering networks yields a misleading perspective on the relations between and risks induced by participating financial institutions; (ii) a multi-layer approach produces a connective structure consistent with most real-world networks (e.g. sparse, inhomogeneous, and clustered); and (iii) the multi-layer network is a multiplex that preserves the main connective features of its constituent layers due to positively correlated multiplexity. The results highlight the importance of mapping and understanding how financial institutions relate to each other across multiple financial environments, and the value of financial market infrastructures as sources of data that may help to overcome the main obstacles for working on multi-layer financial networks.

Evidence from the main Colombian payment and settlement systems verifies that local financial networks have self-organized into a modular (i.e. clustered) scale-free (i.e. inhomogeneous) architecture that favors everyday robustness and performance in exchange for rare episodes of fragility but rapid evolution. Results concur with other real-world networks, and propose new insights and challenges for authorities contributing to financial stability. For instance, (i) traditional reductionist assumptions for modeling financial systems (e.g. homogeneity) may seriously mislead financial authorities’ efforts; (ii) the observed modular scale-free architecture tends to limit cascades and isolate feedbacks; (iii) carelessly reducing inhomogeneity by simply downsizing or dismantling systemically important financial institutions may backfire in the form of a less robust and more crisis-prone financial system; and (iv) financial authorities should understand and take advantage of the existing architecture by means of designing and implementing macro-prudential regulation and system-calibrated requirements.

We model the allocation of central bank liquidity among the participants of the interbank market by using network analysis’ metrics. Our analytical framework considers that a super-spreader simultaneously excels at receiving (borrowing) and distributing (lending) central bank’s liquidity for the whole network, as measured by financial institutions’ hub centrality and authority centrality, respectively. Evidence suggests that the Colombian interbank funds market exhibits an inhomogeneous and hierarchical network structure, akin to a core-periphery organization, in which a few financial institutions fulfill the role of central bank’s liquidity super-spreaders. Our results concur with evidence from other interbank markets and other financial networks regarding the flaws of traditional direct financial contagion models based on homogeneous and non-hierarchical networks. Also, concurrent with literature on lending relationships in interbank markets, we confirm that the probability of being a super-spreader is mainly determined by financial institutions’ size. We provide additional elements for the implementation of monetary policy and for safeguarding financial stability.

Since correlation may be interpreted as a measure of the influence across time‐series, it may be conveniently mapped into a distance and into a weighted adjacency matrix. Based on such matrix, network theory has attempted to filter out the noise in correlation matrices by extracting the dominant hierarchy (i.e. the strongest linear‐dependence signals) within time‐series.The aim of this brief paper is to find the current hierarchy in the sovereigns’ CDS market after the structural shift caused by the failure of Lehman Brothers. Thus, based on two different correlation‐into‐distance mapping techniques and a minimal spanning tree‐based correlation‐filtering methodology on 36 sovereign CDS spread time‐series, the target is to identify which sovereigns are providing the strongest – less noisy – and most informative signals.The resulting sovereigns’ CDS market hierarchy agrees with prior findings of Gilmore et al. (2010) regarding sovereigns’ bonds market, such as the importance of geographical clustering and the idiosyncratic nature of Japan and United States. Additionally, results (i) confirm that a small set of common factors affect the entire system; (ii) identify the relevance of credit rating clustering; (iii) identify Russia, Turkey and Brazil as regional benchmarks; (iv) suggest that lower medium grade rated sovereigns are the most influential, but also the most prone to contagion; and (v) suggest the existence of a Latin American common factor.”

As a result of the most recent global financial crisis, literature has embraced size, connectedness and substitutability as key indicators for financial institutions’ systemic importance. This paper applies Principal Components Analysis to some metrics for assessing size, connectedness and substitutability, where those metrics rely on a combination of balance sheet data and the application of network theory to large-value payment system’s information. The main results are the following: (i) the three concepts and their metrics are explanatory and non-redundant for differentiating financial institutions’ relative systemic importance; (ii) these metrics allow the construction a PCA-based Systemic Importance Index, a valuable tool for financial authorities’ policy and decision-making; and (iii the importance of the too-connected-to-fail criteria and the presence of non-banking firms among the most systemically important financial institutions in the Colombian case are confirmed.

Financial basics and intuition stresses the importance of investment horizon for risk management and asset allocation. However, the beta parameter of the Capital Asset Pricing Model (CAPM) is invariant to the holding period. Such contradiction is due to the assumption of long-term independence of financial returns; an assumption that has been proven erroneous.Following concerns regarding the impact of the long-term dependence assumption on risk (Holton, 1992), this paper quantifies and fixes the CAPM’s bias resulting from this abiding – but flawed – assumption. The proposed procedure is based on Greene and Fielitz (1980) seminal work on the application of fractional Brownian motion to CAPM, and on a revised technique for estimating time-series’ fractal dimension with the Hurst exponent (Leon and Vivas, 2010; Leon and Reveiz, 2011a).Using a set of 85 stocks from the SP on average, the bias is about

Informational constraints may turn the Merton Model for corporate credit risk impractical. Applying this framework to the Colombian financial sector is limited to four stock-market-listed firms; more than a hundred banking and non-banking firms are not listed.Within the same framework, firms’ debt spread over the risk-free rate may be considered as the market value of the sold put option that makes risky debt trade below default-risk-free debt. In this sense, under some supplementary but reasonable assumptions, this paper uses money market spreads implicit in sell/buy backs to infer default probabilities for local financial firms. Results comprise a richer set of (38) banking and non-banking firms. As expected, default probabilities are non-negligible, where the ratio of default-probability-to-leverage is lower for firms with access to lender-of-last-resort facilities.The approach is valuable since it allows for inferring forward-looking default probabilities in the absence of stock prices. Yet, two issues may limit the validity of results to serial and crosssection analysis: overvaluation of default probabilities due to (i) spreads containing non-credit risk factors, and (ii) systematic undervaluation of the firm’s value. However, cross-section assessments of default probabilities within a wider range of firms are vital for financial authorities’ decision making, and represent a major improvement in the implementation of the Merton Model in absence of equity market data.

The most recent financial crisis unveiled that liquidity risk is far more important and intricate than regulation have conceived. The shift from bank-based to market-based financial systems and from Deferred Net Systems to liquidity-demanding Real-Time Gross Settlement of payments explains some of the shortcomings of traditional liquidity risk management.Although liquidity regulations do exist, they still are in an early stage of development and discussion. Moreover, no all connotations of liquidity are equally addressed. Unlike market and funding liquidity, intraday liquidity has been absent from financial regulation, and has appeared only recently, after the crisis.This paper addresses the measurement of Large-Value Payment System’s intraday liquidity risk. Based on the generation of bivariate Poisson random numbers for simulating the minute-by-minute arrival of received and executed payments, each financial institution’s intraday payments time-varying volume and degree of synchrony (i.e. timing) is modeled.To model intraday payments’ uncertainty allows for overseeing participants’ intraday behavior; assessing their ability to fulfill intraday payments at a certain confidence level; identifying participants non-resilient to changes in payments’ timing mismatches; estimating intraday liquidity buffers. Vis-a-vis the increasing importance of liquidity risk as a source of systemic risk, and the recent regulatory amendments, results are useful for financial authorities and institutions.

As a natural extension to Leon and Vivas (2010) and Leon and Reveiz (2010) this paper briefly describes the Cholesky method for simulating Geometric Brownian Motion processes with long term dependence, also referred as Fractional Geometric Brownian Motion (FBM). Results show that this method generates random numbers capable of replicating independent, persistent or antipersistent time‐series depending on the value of the chosen Hurst exponent. Simulating FBM via the Cholesky method is (i) convenient since it grants the ability to replicate intense and enduring returns, which allows for reproducing well‐documented financial returns’ slow convergence in distribution to a Gaussian law, and (ii) straightforward since it takes advantage of the Gaussian distribution ability to express a broad type of stochastic processes by changing how volatility behaves with respect to the time horizon. However, Cholesky method is computationally demanding, which may be its main drawback. Potential applications of FBM simulation include market, credit and liquidity risk models, option valuation techniques, portfolio optimization models and payments systems dynamics. All can benefit from the availability of a stochastic process that provides the ability to explicitly model how volatility behaves with respect to the time horizon in order to simulate severe and sustained price and quantity changes. These applications are more pertinent than ever because of the consensus regarding the limitations of customary models for valuation, risk and asset allocation after the most recent episode of global financial crisis.

The most recent episode of market turmoil exposed the limitations resulting from the traditional focus on too-big-to-fail institutions within an increasingly systemic-crisis-prone financial system, and encouraged the appearance of the too-connected-to-fail (TCTF) concept. The TCTF concept conveniently broadens the base of potential destabilizing institutions beyond the traditional banking-focused approach to systemic risk, but requires methodologies capable of coping with complex, cross-dependent, context-dependent and non-linear systems.After comprehensively introducing the rise of the TCTF concept, this paper presents a robust, parsimonious and powerful approach to identifying and assessing systemic risk within payments systems, and proposes some analytical routes for assessing financial authorities’ challenges. Banco de la Republica’s approach is based on a convenient mixture of network topology basics for identifying central institutions, and payments systems simulation techniques for quantifying the potential consequences of central institutions failing within Colombian large-value payments systems.Unlike econometrics or network topology alone, results consist of a rich set of quantitative outcomes that capture the complexity, cross-dependency, context-dependency and nonlinearity of payments systems, but conveniently disaggregated and dollar-denominated. These outcomes and the proposed analysis provide practical information for enhanced policy and decision-making, where the ability to measure each institution’s contribution to systemic risk may assist financial authorities in their task to achieve payments system’s stability.

Defining whether a financial institution is systemically important (or not) is challenging due to (i) the inevitability of combining complex importance criteria such as institutions’ size, connectedness and substitutability; (ii) the ambiguity of what an appropriate threshold for those criteria may be; and (iii) the involvement of expert knowledge as a key input for combining those criteria. The proposed method, a Fuzzy Logic Inference System, uses four key systemic importance indicators that capture institutions’ size, connectedness and substitutability, and a convenient deconstruction of expert knowledge to obtain a Systemic Importance Index.This method allows for combining dissimilar concepts in a non-linear, consistent and intuitive manner, whilst considering them as continuous –non binary- functions. Results reveal that the method imitates the way experts them-selves think about the decision process regarding what a systemically important financial institution is within the financial system under analysis.The Index is a comprehensive relative assessment of each financial institution’s systemic importance. It may serve financial authorities as a quantitative tool for focusing their attention and resources where the severity resulting from an institution failing or near-failing is estimated to be the greatest. It may also serve for enhanced policy-making (e.g. prudential regulation, oversight and supervision) and decision making (e.g. resolving, restructuring or providing emergency liquidity).

Despite foreign reserves’ strategic asset allocation relies mainly on Modern Portfolio Theory (MPT), the unique characteristics of central banks obliges them to articulate and reconcile typical optimization procedures with reserves’ management objectives such as providing confidence regarding the ability to meet the country’s external commitments. Moreover, further involvedness come from broad economic factors as diverse as the openness of capital and current accounts, external debt’s maturity and currency composition, and exchange rate regime. Therefore, in order to alleviate the divergence from theory and practice regarding foreign reserves’ strategic asset allocation, this paper describes the methodologies and procedures developed and employed by the Foreign Reserves Department of Banco de la República. The mainstay of the paper is a long-term-dependence-adjusted and non-loss-constrained version of the Black-Litterman model for obtaining the efficient frontier from a set of investments complying with safety, liquidity and return criteria, where the choice of the portfolio which maximizes utility makes use of an estimation of the Board of Directors’ risk aversion. Results exhibit the effects of the unique nature of foreign reserves management for emerging markets. Typical features of foreign reserves management by central banks, such as non-loss restrictions due to capital preservation objectives, result in increased complexity in the optimization process and in asset allocations significantly distant from standard MPT’s optimality.

Este documento estudia la estabilidad del sistema de pagos (SP) de alto valor en Colombia (CUD) ante el incumplimiento de una entidad sistemicamente importante, y evalua la capacidad de respuesta de las entidades afectadas a partir de la utilizacion de sus recursos y a traves de los mecanismos de liquidez que brinda el Banco de la Republica. De acuerdo con la literatura reciente, las entidades sistemicamente importantes se identifican bajo el concepto de too-connected-to-fail (TCTF) para diferentes escenarios de volatilidad del mercado de TES y de actividad del SP. La estabilidad del SP se evalua mediante Topologia de Redes (TR) y un Modelo de Simulacion de Pagos (MSP), el cual incorpora un algoritmo de resolucion de colas recursivo tipo FIFO (First In First Out) y un algoritmo de compensacion multilateral. Los resultados de la TR sugieren que el CUD es una red de tamano mediano, robusta, estable y concentrada. El MSP mostro, ademas, que variables como los saldos de las entidades en el CUD, la oportunidad de las transacciones intradia, y la concentracion de liquidez, inciden sobre el numero de entidades afectadas.Se encontro que la mayoria de las entidades cuenta con mecanismos que les permiten solventar la iliquidez temporal en el SP. Sin embargo, existen entidades que, por su estructura y especialidad de su negocio, deben hacer un mayor esfuerzo en la administracion del riesgo de liquidez.

Es una practica muy difundida el multiplicar la desviacion estandar por la raiz del tiempo para escalarla a otros plazos. Asi, con base en la estimacion de la desviacion estandar o del VaR (Value at Risk) diario, es usual obtener la desviacion estandar o el VaR para un periodo de diez dias como el producto de la primera y la raiz de diez. Esta practica, basada en la hipotesis de los mercados eficientes, supone que los cambios de los precios son independientes entre si; es decir, que las series de tiempo no presentan memoria. El presente documento se ocupa de estimar la presencia de memoria de largo plazo en los mercados cambiario, accionario y de renta fija colombianos, para lo cual se utiliza la metodologia de rango reescalado clasico (R/S) y modificado (mR/S). Ademas de encontrar persistencia significativa para el mercado accionario y de renta fija, se estima el exponente de Hurst ajustado, el cual sirve para cuantificar el error derivado de algunas practicas basadas en el supuesto de independencia. De acuerdo con los resultados obtenidos para dichos mercados, (i) el supuesto segun el cual los precios reflejan toda la informacion disponible es errado; (ii) algunas practicas en la optimizacion de portafolios y la valoracion de activos son cuestionables, y (iii) tal como se infiere de la revision hecha en 2009 por el Comite de Basilea a los estandares cuantitativos para el calculo del riesgo de mercado, la regla de la raiz del tiempo subestima significativamente el riesgo.

Operational Risk (OR) results from endogenous and exogenous risk factors, as diverse and complex to assess as human resources and technology, which may not be properly measured using traditional quantitative approaches. Engineering has faced the same challenges when designing practical solutions to complex multifactor and non-linear systems where human reasoning, expert knowledge or imprecise information are valuable inputs. One of the solutions provided by engineering is a Fuzzy Logic Inference System (FLIS). Despite the goal of the FLIS model for OR is its assessment, it is not an end in itself. The choice of a FLIS results in a convenient and sound use of qualitative and quantitative inputs, capable of effectively articulating risk management's identification, assessment, monitoring and mitigation stages. Different from traditional approaches, the proposed model allows evaluating mitigation efforts ex-ante, thus avoiding concealed OR sources from system complexity build-up and optimizing risk management resources. Furthermore, because the model contrasts effective with expected OR data, it is able to constantly validate its outcome, recognize environment shifts and issue warning signals.

First developed by Markowitz (1952), the mean-variance framework is the most widespread theoretical approximation to the portfolio problem. Nevertheless, successful application in the investment community has been limited. Assumptions such as normality of returns and a static correlation matrix could partially account for this. To overcome some of the limitations of the mean-variance framework, mainly the choice of the risk metric and the inconvenience of using an estimated correlation matrix typical of tranquil or euphoria periods, this paper proposes an alternative risk measure: the maximum drawdown (MDD), and combines it with a wealth creation measure to define a new portfolio optimization space. Like other market practitioners’ measures, MDD lacks of a complete and solid theoretical foundation. In an effort to contribute to its theoretical foundation, this paper uses common sense and financial intuition to introduce such measure, followed by a review of its technical advantages and coherence for risk management. Finally, an application of a MDD risk metric based portfolio optimization model is presented. The main findings indicate this proposal may effectively help overcome some of the traditional mean-variance shortcomings and provide some useful tools for portfolio optimization in practice. For long-term performance driven portfolios, such as pension funds, this approach may yield interesting results because it focuses on wealth creation over the long run.