Academic Commons

Theses Doctoral

Essays on the Effects of Frictions on Financial Intermediation

Bolandnazar, Mohammadreza

This dissertation aims to study the behavior of intermediaries under market imperfections and the consequences of that for the financial market's functioning. To do so, I focus on two classes of market frictions: funding constraints and information asymmetry. Chapter 1 studies how the dealers' capital constraints affect the market liquidity in the presence of imperfect competition and how recent regulations have shifted the competitive landscape of interest rate swaps. On the subject of informational frictions, Chapters 2 and 3 study empirically and theoretically the pace at which prices incorporate private information under the limited learning capacity of the informed traders.

Understanding the microstructure of the swap markets is of interest to both policymakers and academics, especially for it helps in the efficient implementation of post-crisis regulations, namely the Dodd-Frank Act. An understudied dimension of the swap market microstructure is the determinants of the cost of the market-making activity. Using a proprietary regulatory dataset collected by the Commodity Futures Trading Commission (CFTC) on both the interest rate swap transactions and the collateral requirements at the London Clearinghouse (LCH), in Chapter 1, I study the key balance sheet constraints that affect the ability of the bank-affiliated dealers to provide intermediation service to the end-users. Most of the interest rate swaps are now mandated to be centrally cleared. This has increased the dealer's need for collateral in the form of highly liquid assets (cash and cash equivalents) to back their swap exposures. Facing capital adequacy measures such as Supplementary Leverage Ratio (SLR), dealers find it even costlier to increase the size of their balance sheet to fund these margins.

I show that a 1-percentage point increase in SLR leads to an increase of 1.09 percentage points in the bank's cost of capital per unit of margin requirement. Furthermore, I find the funding spread of the dealers (the difference between the cost of external funding and the risk-free rate) is also a relevant factor for determining the dealer's marginal cost of swap transaction; a cost that is evidently transferred to the end-users in the form of less favorable prices. Measuring the cost of intermediation for the dealer-to-client interest rate swap market is challenging because of the high concentration in the market-- the first seven dealers intermediate more than 50% of the total notional traded. Therefore, one must consider the nontrivial effect of markups in transaction prices to estimate the marginal cost of intermediation reliably. For this reason, I model a differentiated product demand for swaps in the spirit of empirical Industrial Organization (IO) literature and structurally estimate this model to account for the markups in the transaction prices using estimated price elasticities. The demand estimations show economically interpretable heterogeneity among the end-users in their taste for duration risk hedging. The structurally estimated equilibrium model of intermediation can serve as a basis for answering counterfactual policy questions, especially in the debate on the social costs and benefits of excluding initial margins in calculating supplementary leverage ratio.

In Chapter 2, I turn the focus to the impact of informational frictions on market-making activity. More specifically, we study the informed trading under random stopping time. Empirical evidence is provided based on an episode of time when the Securities and Exchange Commission (SEC) unintentionally disclosed security filings to some investors before the public for several years. For technological reasons, the delay between the private and public disclosure was exogenously random. We exploit the variation in the time window of private information to show the intensity of trades and the speed at which market prices reach their efficiency, decrease with the expected arrival time of public announcement. In addition, we find the learning capacity of the insider determines the evolution of trading intensity over time.

In Chapter 3, inspired by the stylized facts observed in the earlier chapter, I extend the Kyle (1985) model of strategic trading to a case with limited learning capacity of both the dealers and the informed traders (insiders). The insider does not perfectly observe the true value of the security, but he continues to hone his knowledge by using private information sources over time. Two classes of equilibria emerge from this model. In one class, the insider trades excessively patiently, and the market efficiency is reached only asymptotically. In the second type, the insider optimally chooses a deterministic time T, before which he trades patiently as in Kyle (1985) until the price reaches its full efficiency. After T, the insider keeps revealing every piece of new information immediately, and the market price stays efficient while the insider keeps making profits. Which equilibrium emerges depends on the insider's learning capacity, initial informational advantage, and the private source's informational content.


  • thumnail for Bolandnazar_columbia_0054D_16685.pdf Bolandnazar_columbia_0054D_16685.pdf application/pdf 1.25 MB Download File

More About This Work

Academic Units
Thesis Advisors
Johannes, Michael S.
Ph.D., Columbia University
Published Here
July 16, 2021