Ebook Crises and Liquidity in Over-the-Counter Markets
A crucial aspect of the recent financial turmoil has been the severe liquidity dry-up in over-the-counter (OTC) markets, and its role in the unfolding of the crisis. Many of the financial instruments at the core of the crisis mortgage backed securities, collateralized debt obligations, credit default swaps are indeed traded over-the-counter through bilateral trades, outside of organized exchanges. Liquidity in those markets is provided on a voluntary basis by brokers-dealers, such as large investment banks. While liquidity provision seems inconspicuous in normal times, it has proved vulnerable recently, as revealed by the rapid drying-up of liquidity in markets for credit derivatives, commercial paper and municipal bonds alike.
The lack of resilience of OTC markets has been attributed to the inadequate regulation of these markets, as well as to the absence of “designated” marketmakers or dealers. The lax regulations, it is argued, may have led dealers to commit too little marketmaking capital. In addition, even if dealers are well capitalized in normal times, they may have their credit lines severed during a crisis, and effectively withdraw from market marking. In extreme cases, they can even go bankrupt, as shown by the collapse of Lehman Brothers, one of the most active participants in OTC derivatives markets.
Analyzing the policies that aim at restoring liquidity in OTC market requires a framework describing explicitly the frictions and trading mechanisms in such markets. The objective of this paper is to propose such a model of liquidity provision in OTC markets and to study its implications for the functioning of markets and their allocative efficiency in times of crisis. A key insight of our analysis is to show that in some very illiquid and opaque markets, even well capitalized dealers may fail to provide liquidity when it would be socially efficient to do so. Injecting more capital would be ineffective because dealers would hoard the added “funding liquidity” instead of purchasing outside investors’ assets on their own account.
We show that, in order for capital injections to be effective, they should be combined with market reforms that facilitate trades (e.g., by promoting information disclosure, standardization of assets, electronic trading platforms). For instance, the difficulty to trade in OTC markets may be the symptom of some underlying adverse selection problem induced by unobserved counterparty risk. Reducing this friction could mean, for instance, setting a Central counterparty Clearing House (CCP), a new financially sound intermediary, providing safe collateral, or letting a public agency purchase assets, with a profit maximizing objective. Similarly, our model recommends combining capital injections with market reforms that erodes dealers’ market power which could be achieved, for instance, by promoting greater price transparency.
The market setting we consider is based on the search-theoretic model of OTC markets by Duffie, Gârleanu, and Pedersen (2005) (DGP hereafter) and its extensions by Weill (2007) and Lagos and Rocheteau (2007, 2009). As in these models, we assume that outside investors cannot trade continuously in a centralized Walrasian market. Instead, they receive infrequent and random trading opportunities with intermediaries called “dealers,” who are able to trade continuously with each others. While this search friction provides a natural description of bilateral trades in OTC markets, it also applies more broadly: it captures a wide range of frictions that make it more difficult to trade financial assets during crises such as disruptions in communication systems, adverse selection problems with distressed intermediaries or outright dealers’ failures. To create a crisis, we hit our search market with an aggregate negative shock that reduces investors’ willingness to hold the asset. The crisis persists until some random time at which point investors receive the opposite shock and the economy recovers. Then, we determine the conditions under which well-capitalized dealers provide liquidity to outside investors, accumulating assets in their inventory during the crisis and unloading these assets when the economy recover.
Since our dealers are profit maximizing, their incentives to provide liquidity are driven by anticipated capital gains. Therefore, dealers are more likely to accumulate inventories if the initial price drop is large and followed by a quick rebound: i.e. if the crisis is severe enough and anticipated to be short lived. We find that the amount of liquidity provided by well-capitalized dealers varies non-monotonically with the magnitude of two trading frictions: the search friction and the market power of dealers. Precisely, consider a spectrum of OTC markets ranging from those with very small frictions, for instance markets for Treasury securities or wholesale foreign exchange, to those with large trading frictions, such as some markets for subprime mortgage backed securities.
Then, under natural conditions, we find that dealers provide no liquidity in markets at both end of the spectrum, and some liquidity in markets lying in the the middle of the spectrum. When frictions are small, investors choose to take more extreme positions because they know that they can re-balance their asset holdings very quickly. Some investors supply so much liquidity to other investors, that dealers don’t find it profitable to step in. At the other end of the spectrum when trading frictions are very large, investors become reluctant to hold extreme asset positions, because they anticipate these will be very difficult to unwind. In an equilibrium, all investors end up with a similar “average” asset position, do not re-balance much, and therefore do not demand much liquidity from dealers. We show that, because of this endogenous negative shift in liquidity demand, in equilibrium dealers do not provide any liquidity. As mentioned above, the main policy implication of this finding is that injecting capital may prove ineffective in markets with very large friction: dealers would hoard it and would not provide liquidity by accumulating asset inventories.
Our work belong to the recent literature that studies trading frictions in asset markets. DGP focused on steady states and their analysis is silent about liquidity provision by dealers. Weill (2007) studies the timing of liquidity provision by dealers along the transitional path in a dynamic version of DGP. Weill and the literature spurred by DGP, however, keep the framework tractable by imposing a restriction on asset holdings, namely, that investors can only hold either 0 or 1 unit of the asset. Lagos and Rocheteau (2007, 2009) study a version of DGP where investors can hold unrestricted asset positions, creating a non-trivial extensive margin in investors’ demand for liquidity. In this paper, we go beyond previous studies by studying the out-of-steady-state dynamics induced by aggregate shocks while allowing both dealers and investors to hold unrestricted asset positions.
This generates new implications for the demand and the supply of liquidity that are relevant for the study of crisis. For instance we find that, in contrast to Weill (2007), even well-capitalized dealers may sometimes not find it in their interest to provide liquidity during the crisis when it would be socially optimal to do so, implying that capital injection may be ineffective. Weill (2007) also assumed that, after the initial shock, the recovery path was deterministic. While this is a useful simplifying assumption, it is clearly at odd with the considerable uncertainty faced by market participant during crises. We go beyond this analysis by making the recovery a random event. This makes the model more realistic and generates the new implication that our rational dealers find it optimal to buy asset while the market price continues to decline, and re-sell them while the market price continues to go up.
Download
PDF Ebook Crises and Liquidity in Over-the-Counter Markets
Posted in :
