Jean-Edouard Colliard

Monitoring the Supervisors: Optimal Regulatory Architecture in a Banking Union

I study the optimal architecture of bank supervision in a federal system. A central supervisor gets information about a bank, for instance through stress-testing, and decides whether an on-site examination should be performed by a local or a central authority. Local supervisors have lower inspection costs, but do not internalize cross-border externalities. The optimal degree of centralization depends on the severity of these externalities, the opacity of the supervised bank and the specificity of its assets. The market reacts to the chosen architecture, so that a centralized supervision endogenously increases market integration and cross-border externalities, strengthening the need for centralized supervision. The economy can be trapped in an equilibrium with low supervision and integration, while a forward-looking design of the supervisory architecture would coordinate economic agents on a superior equilibrium.

Financial Transaction Taxes, Market Composition, and Liquidity

We use the introduction of a financial transaction tax (FTT) in France in 2012 to test competing theories on its impact. We find no support for the idea that an FTT improves market quality by affecting the composition of trading volume. Instead, our results are in line with the hypothesis that a lower trading volume reduces liquidity, and thereby market quality. Consistent with theories of asset pricing under transaction costs, we document a shift in security holdings from short-term to long-term investors. Finally, our findings show that moderate aggregate effects on market quality can mask large adjustments made by individual agents. JEL Classification: G10, G14, G18, H32This paper presents evidence on the causal impact of nancial transaction taxes in a fully electronic market. Our dierence-in-dier ences approach examines the trading activity in French stocks around the introduction of an FTT together with a local HFT tax on August 1st, 2012, relative to a control group of comparable Dutch stocks traded under identical rules. We show that the policy change has led to a large ( 30%) temporary decrease in trading volume during August, which surprisingly was not associated with a signicant deterioration of market quality, with the exception of market depth. Most of the impact on volume is reversed in September. In contrast to this largely transitory eect, we uncover a more permanent impact on low-latency activity, as indicated by less aggressive orders and fewer rapid cancellations. Our results suggest that not all types of market activity seem to contribute equally to market quality.

Optimal Supervisory Architecture and Financial Integration in a Banking Union

Both in the United States and in the Euro area, bank supervision is the joint responsibility of local and central/federal supervisors. I study how such a system can optimally balance the lower inspection costs of local supervisors with the ability of the central level to internalize cross-border or interstate externalities. The model predicts that centralised supervision endogenously increases market integration and cross-border externalities, further strengthening the need for centralised supervision. This complementarity implies that, for some parameterizations of the model, the economy can be trapped in a local supervision equilibrium with low supervision and integration. In such a case, the forward-looking introduction of a centralized supervisory architecture achieves a superior equilibrium. JEL Classification: D53, G21, G28, G33, G38, L51

Multinational Banks and Supranational Supervision

Supervision of multinational banks (MNBs) by national supervisors suffers from coordination failures. We show that supranational supervision solves this problem, and decreases the expected costs of a MNBs default, taking its organizational structure as given. However, the MNB strategically adjusts its structure to the new supervisory framework. It converts its subsidiary into a branch, or conversely, with a view to reducing supervisory monitoring. We identify the cases in which this endogenous reaction leads to unintended consequences, such as higher costs to the deposit insurance fund, lower welfare, or closure of the MNBs foreign unit. Current reforms of MNB supervision should thus take into account that MNBs adapt their organizational structures to changes in supervision.

Strategic Selection of Risk Models and Bank Capital Regulation

The regulation of banks uses their internal risk estimates to compute finer capital requirements. The underlying risk models have been blamed for their optimism, which I link to a hidden information problem between a regulator and a bank better informed about risk models. I first show that low incentives to use cautious models can seriously cripple current reforms: a regulatory tightening to compensate for “model risk” (switching to Basel III/increasing a floor on capital ratios) leads to a contraction of credit, which favors a wider adoption of optimistic models, increases risk and can make counter-cyclical buffers pro-cyclical. Second, giving proper incentives is difficult: as model uncertainty relates to tail risk, over-optimism is typically revealed only when it is too late to impose a penalty. The framework allows to derive predictions on the use of internal models and to compare different regulatory solutions. More broadly, this paper shows how incentives impact the development of applied economic models. Journal of Economic Literature Classification Number : D82, D84, G21, G32, G38.

Catching Falling Knives: Speculating on Liquidity Shocks

Many market participants invest resources to acquire information about liquidity rather than fundamentals. I show that agents using such information can reduce the magnitude of short-lived pricing errors by trading against liquidity shocks. However, the short-run stabilizing effect of this behavior also makes it more difficult to identify liquidity shocks, a signal-jamming effect that slows down price discovery in the long-run. As more agents invest in non-fundamental information, market prices become more resilient to liquidity shocks, but also recover more slowly from temporary price deviations.