Final Report Summary - FRICTIONS (Frictions in the Financial System)
The financial crisis in 2008 and the subsequent European debt crisis demonstrated that global growth relies heavily on well-functioning financial markets. Fractures in these markets can paralyze even the most developed economies of the world. Still, we know little on the exact mechanisms of how small frictions in financial markets turn to large fractures in the real economy and lead to crises. The main objective of my ERC research was to understand each step of this phenomenon better. I followed a dual approach to achieve this goal.
I. The Micro Approach: Which are the relevant frictions in financial markets?
First, it is crucial to identify the relevant frictions hindering the efficient functioning of specific markets. Each project in this group following this approach zooms in to a specific market segment and highlights the effects of a given friction in that market.
In particular, Babus and Kondor (2018) and Kondor and Pinter (2018) are focusing on information frictions in decentralized markets using network technics, Kondor and Zawadowski (2018) focuses on the effect of learning to the efficiency of the aggregate capital reallocation process and welfare, while Kondor and Koszegi (2017) focuses on financial products exploiting households biases in information processing. All three projects studies problems of imperfect information. However, the three projects focus on different market segments and use different methodological approaches. I believe this way we can get closer to a full picture.
In the first of our network project, “Trading and information diffusion in over-the-counter markets” (with Ana Babus), we focus on over-the-counter (OTC) markets. In these markets, transactions are bilateral, prices are dispersed, trading relationships are persistent, and typically, a few large dealers intermediate a large share of the trading volume. In this paper, we explore a novel approach to modelling OTC markets that reflects these features. In our model, each dealer with private information can engage in several bilateral transactions with her potential trading partners, as determined by her links in a network. Each dealer’s strategy is represented as a quantity-price schedule. Our focus is on how decentralization (characterized by the structure of the dealer network) and adverse selection jointly influence information diffusion, expected profits, trading costs, and welfare. We prove that information diffusion through prices is not affected by strategic considerations in a well-defined sense. We show that each equilibrium price depends on all the information available in the economy, incorporating even the signals of dealers located far from a given transaction. We identify an informational externality that constrains the informativeness of prices. We highlight that decentralization can both increase or decrease welfare and that an important determinant of a dealer’s trading cost besides her own centrality is the centrality of her counterparties. Using an example calibrated to securitization markets, we argue that in realistic interdealer networks, more central dealers learn more, trade more at lower costs, and earn higher expected profit. However, we also explain why in some special cases, more-connected dealers can earn a lower expected profit.
In our follow-up empirical project, “Private Information and Client Centrality in the UK Gilt Market’’, (with Gábor Pintér) we also study informational frictions and financial intuitions. However, in this project, with the help of a proprietary data-set we turn to the treasury market. Treasury markets have a crucial role in the financial infrastructure, determining the yield curve, affecting the cost of financing across the economy and where monetary policy is implemented. Also, it has an over-the-counter structure, where clients, like hedge funds, commercial banks, and foreign central banks, trade bilaterally with a small group of primary dealers. As a starting point, we have observed that the number of dealers a given client uses to trade varies a lot from month to month. Our hypothesis is that clients only occasionally trade on private information, and when they do, they prefer to trade with more dealers to hide this information from the market. Indeed, our results show that if a client buys a security in a month when he is more connected, that security tends to perform better up to 10 days than those securities he buys when he is less connected. That is, time-variation in connectedness makes private information -- an object notoriously hard to measure -- measurable. With this proxy, we can turn to understand the nature of private information in this market. Our results suggest that this information helps clients to predict the trade flows of other clients in subsequent days. More interestingly, this private information seems to be disseminated by the client’s own dealer.
In our project “Financial Information and Financial Choice” (with Botond Koszegi), we analyze the implications of increases in the selection of, and information about, derivative financial products in a model in which investors neglect informational differences between themselves and issuers. We assume that investors receive information that is noisy and inferior to issuers’ information, and that issuers can select the set of underlying assets when designing a security. In contrast to the received wisdom that diversification is helpful, we show that when custom-designed diversification across a large number of underlying assets is possible, then expected utility approaches negative infinity. Even beyond this limiting case, any expansion in choice induced by either an increase in the maximum number of assets underlying a security, or an increase in the number of assets from which the underlying can be selected, Pareto-lowers welfare. Furthermore, under reasonable conditions an improvement in investor information Pareto-lowers welfare by giving investors the false impression that they can spot good deals. An increase in competition between issuers does not increase welfare, and even increases investors’ incentive to acquire welfare reducing information. Restricting the set of underlying assets the issuer can use—a kind of standardization—raises welfare, and once this policy is adopted, increasing investor information becomes beneficial.
II. The Macro Approach: Aggregate Liquidity Fluctuations and Global Cycles
In Kondor and Vayanos (2018) and Farboodi and Kondor (2018), I focused on how particular type of frictions affect capital supply, and as a result, the real economy. As these are traditionally macro topics, I call this the macro approach.
In the project “Liquidity risk and the dynamics of arbitrage capital” (with Dimitri Vayanos) we study time-variation in the risk-bearing capacity of the financial sector. Liquidity in financial markets is often provided by specialized agents, such as market makers, trading desks in investment banks and hedge funds. Adverse shocks to the capital of these agents cause liquidity to decline and risk premia to increase. Conversely, movements in the prices of assets held by liquidity providers feed back into these agents’ capital. In this paper we study the dynamics of liquidity providers’ capital, the liquidity that these agents provide to other participants and assets’ risk premia. We build a framework with minimal frictions, in particular no asymmetric information or borrowing constraints. The capital of liquidity providers matters in our model only because of standard wealth effects. At the same time, we depart from most frictionless asset-pricing models by fixing the riskless rate and by suppressing wealth effects for agents other than the liquidity providers. These assumptions are sensible when focusing on shocks to the capital of liquidity providers in an asset class rather than in the entire asset universe. Our combination of assumptions makes it possible to prove general analytical results on equilibrium prices and allocations. We characterize, in particular, how liquidity providers’ risk-appetite, the endogenous risk that they generate, and the pricing of that risk depend on liquidity demanders’ characteristics and on liquidity providers’ capital. We also show that the capital of liquidity providers is the single priced risk factor, and that liquidity aggregated over the assets that we consider captures that factor because it increases in capital. Our results thus suggest that a priced liquidity risk factor may arise even with minimal frictions.
In our project “Heterogeneous Global Cycles” (with Maryam Farboodi), we study why countries differ in terms of their exposure to fluctuations in the global supply of credit. We argue that frictions in global intermediation lead to an endogenous partitioning of economies into groups with low and high exposure to the global credit cycle. We show that investors with varying degree of information hold dissimilar portfolios, with low skilled investors sharply rebalancing their cross-country asset holdings across different aggregate states. The differential response of investors invites differential strategies of firms, jointly shaping heterogeneous global cycles. We connect the implications of our model to stylized facts on credit spreads, investment, safe asset supply, concentration of debt ownership, and the return on debt during various boom-bust episodes, both in the time series and in the cross-section. We demonstrate that a global savings glut not only exacerbates both booms and busts in high exposure countries, but also increases the exposure of some countries to credit cycles.
I. The Micro Approach: Which are the relevant frictions in financial markets?
First, it is crucial to identify the relevant frictions hindering the efficient functioning of specific markets. Each project in this group following this approach zooms in to a specific market segment and highlights the effects of a given friction in that market.
In particular, Babus and Kondor (2018) and Kondor and Pinter (2018) are focusing on information frictions in decentralized markets using network technics, Kondor and Zawadowski (2018) focuses on the effect of learning to the efficiency of the aggregate capital reallocation process and welfare, while Kondor and Koszegi (2017) focuses on financial products exploiting households biases in information processing. All three projects studies problems of imperfect information. However, the three projects focus on different market segments and use different methodological approaches. I believe this way we can get closer to a full picture.
In the first of our network project, “Trading and information diffusion in over-the-counter markets” (with Ana Babus), we focus on over-the-counter (OTC) markets. In these markets, transactions are bilateral, prices are dispersed, trading relationships are persistent, and typically, a few large dealers intermediate a large share of the trading volume. In this paper, we explore a novel approach to modelling OTC markets that reflects these features. In our model, each dealer with private information can engage in several bilateral transactions with her potential trading partners, as determined by her links in a network. Each dealer’s strategy is represented as a quantity-price schedule. Our focus is on how decentralization (characterized by the structure of the dealer network) and adverse selection jointly influence information diffusion, expected profits, trading costs, and welfare. We prove that information diffusion through prices is not affected by strategic considerations in a well-defined sense. We show that each equilibrium price depends on all the information available in the economy, incorporating even the signals of dealers located far from a given transaction. We identify an informational externality that constrains the informativeness of prices. We highlight that decentralization can both increase or decrease welfare and that an important determinant of a dealer’s trading cost besides her own centrality is the centrality of her counterparties. Using an example calibrated to securitization markets, we argue that in realistic interdealer networks, more central dealers learn more, trade more at lower costs, and earn higher expected profit. However, we also explain why in some special cases, more-connected dealers can earn a lower expected profit.
In our follow-up empirical project, “Private Information and Client Centrality in the UK Gilt Market’’, (with Gábor Pintér) we also study informational frictions and financial intuitions. However, in this project, with the help of a proprietary data-set we turn to the treasury market. Treasury markets have a crucial role in the financial infrastructure, determining the yield curve, affecting the cost of financing across the economy and where monetary policy is implemented. Also, it has an over-the-counter structure, where clients, like hedge funds, commercial banks, and foreign central banks, trade bilaterally with a small group of primary dealers. As a starting point, we have observed that the number of dealers a given client uses to trade varies a lot from month to month. Our hypothesis is that clients only occasionally trade on private information, and when they do, they prefer to trade with more dealers to hide this information from the market. Indeed, our results show that if a client buys a security in a month when he is more connected, that security tends to perform better up to 10 days than those securities he buys when he is less connected. That is, time-variation in connectedness makes private information -- an object notoriously hard to measure -- measurable. With this proxy, we can turn to understand the nature of private information in this market. Our results suggest that this information helps clients to predict the trade flows of other clients in subsequent days. More interestingly, this private information seems to be disseminated by the client’s own dealer.
In our project “Financial Information and Financial Choice” (with Botond Koszegi), we analyze the implications of increases in the selection of, and information about, derivative financial products in a model in which investors neglect informational differences between themselves and issuers. We assume that investors receive information that is noisy and inferior to issuers’ information, and that issuers can select the set of underlying assets when designing a security. In contrast to the received wisdom that diversification is helpful, we show that when custom-designed diversification across a large number of underlying assets is possible, then expected utility approaches negative infinity. Even beyond this limiting case, any expansion in choice induced by either an increase in the maximum number of assets underlying a security, or an increase in the number of assets from which the underlying can be selected, Pareto-lowers welfare. Furthermore, under reasonable conditions an improvement in investor information Pareto-lowers welfare by giving investors the false impression that they can spot good deals. An increase in competition between issuers does not increase welfare, and even increases investors’ incentive to acquire welfare reducing information. Restricting the set of underlying assets the issuer can use—a kind of standardization—raises welfare, and once this policy is adopted, increasing investor information becomes beneficial.
II. The Macro Approach: Aggregate Liquidity Fluctuations and Global Cycles
In Kondor and Vayanos (2018) and Farboodi and Kondor (2018), I focused on how particular type of frictions affect capital supply, and as a result, the real economy. As these are traditionally macro topics, I call this the macro approach.
In the project “Liquidity risk and the dynamics of arbitrage capital” (with Dimitri Vayanos) we study time-variation in the risk-bearing capacity of the financial sector. Liquidity in financial markets is often provided by specialized agents, such as market makers, trading desks in investment banks and hedge funds. Adverse shocks to the capital of these agents cause liquidity to decline and risk premia to increase. Conversely, movements in the prices of assets held by liquidity providers feed back into these agents’ capital. In this paper we study the dynamics of liquidity providers’ capital, the liquidity that these agents provide to other participants and assets’ risk premia. We build a framework with minimal frictions, in particular no asymmetric information or borrowing constraints. The capital of liquidity providers matters in our model only because of standard wealth effects. At the same time, we depart from most frictionless asset-pricing models by fixing the riskless rate and by suppressing wealth effects for agents other than the liquidity providers. These assumptions are sensible when focusing on shocks to the capital of liquidity providers in an asset class rather than in the entire asset universe. Our combination of assumptions makes it possible to prove general analytical results on equilibrium prices and allocations. We characterize, in particular, how liquidity providers’ risk-appetite, the endogenous risk that they generate, and the pricing of that risk depend on liquidity demanders’ characteristics and on liquidity providers’ capital. We also show that the capital of liquidity providers is the single priced risk factor, and that liquidity aggregated over the assets that we consider captures that factor because it increases in capital. Our results thus suggest that a priced liquidity risk factor may arise even with minimal frictions.
In our project “Heterogeneous Global Cycles” (with Maryam Farboodi), we study why countries differ in terms of their exposure to fluctuations in the global supply of credit. We argue that frictions in global intermediation lead to an endogenous partitioning of economies into groups with low and high exposure to the global credit cycle. We show that investors with varying degree of information hold dissimilar portfolios, with low skilled investors sharply rebalancing their cross-country asset holdings across different aggregate states. The differential response of investors invites differential strategies of firms, jointly shaping heterogeneous global cycles. We connect the implications of our model to stylized facts on credit spreads, investment, safe asset supply, concentration of debt ownership, and the return on debt during various boom-bust episodes, both in the time series and in the cross-section. We demonstrate that a global savings glut not only exacerbates both booms and busts in high exposure countries, but also increases the exposure of some countries to credit cycles.