Systemic Risk and the Macroeconomy

Systemic Risk and the Macroeconomy

Discussion of: Capital Requirements in a Quantitative Model of Banking Industry Dynamics by Dean Corbae and Pablo DErasmo (CD) Gianni De Nicol International Monetary Fund and CESifo The views expressed in this presentation are those of the author and do not necessarily represent those of the IMF. CD contribution This is an ambitious and important paper It is (to the best of my knowledge) the first paper to deliver a quantitative model of a banking industry with heterogeneous

banks This is extremely useful for policy evaluation My key comment: point estimates of policy counterfactuals need to be associated with estimates Key questions addressed by recent industry and general equilibrium models of banking What is the quantitative impact of new Basel III capital requirements on: A. Lending (and real activity) B. Welfare Most models have representative banks, no heterogeneity CD introduce bank heterogeneity : this allows them to evaluate the impact of capital requirements on market

structure (entry and exit) and Key features of the CD Model Sequence of Stackelberg -Cournot equilibria, where the endogenous bank size distribution is obtained by adding entry/exit decisions. The model is calibrated to match long-run averages of some banking industry statistics (the success is so far partial, see Table 5). The introduction of bank heterogeneity comes at the unavoidable cost of several simplifications Example: banks are owned by infinitely Quantitative impact of increased capital requirements on lending

a. b. c. Recent equilibrium models generate a broadly consistent qualitative picture of the (long-term) impact of an increase in capital requirements on lending and real activity: Capital requirements reduce bank risk lending and real activity decline there exists an inverted-U shaped relationship between bank lending, welfare, and capital requirements. Yet, quantitative assessments differ substantially. How can we compare these quantitative results? Models are not nested Paper required capital

lending Welfare change De Nicol et al. (2014) 0 to 4 4 to 12 15 2.4 + Martinez-Miera and Suarez (2014) 7 to 14 20 + Nguyen (2014) 4 to 8

8 to 20 + Clerc et al. (2014) 8 to 10 10 to 14 -1.3 -4.5 + Begenau (2015) 6 to 14 2.1 +

Corbae and DErasmo (2016) 4 to 6 8 The closest model comparable to CD De Nicol, Gamba and Lucchetta ( DNGL, RFS 2014) Banks dynamically transform short term liabilities into longer-term illiquid assets whose returns are uncertain (no maturity transformation in CD, but equilibrium loan rates in CD and not in DNGL) .

Deposits are insured (as in CD) Banks can be in financial distress, can issue debt or equity with issuance costs (as in CD) Financial market equilibrium, endogenous discount factor (not in CD) Pricing of systematic (aggregate) and idiosyncratic risks, large set of states on a truncated normal grid ( 2 aggregate states in How capital and associated liquidity choices move along the business cycle? CD: Loan supply is pro-cyclical (as in DNGL). (DNGL) Under mild capital requirements, banks buildup capital buffers in upturns to reduce the risk of costly loan liquidations in downturns , but ratios decline (as in CD, Figure 18) (DNGL) In a downturn banks use retained earnings and the financial

resources freed by the reduction in lending to strengthen their liquidity position (as in CD, p.39). a. c. An increase in capital requirements: the competitive effects Important CD results: Exit rates decline for fringe banks. The leader bank reduces loan supply and induces entry of small banks, but charter value declines dominate, reducing the size of the fringe bank sector increase in concentration On net: lower loan supply (-8.7%), increase in interest rates (+7.6)% and corporate default frequencies (+12.2%) Yet, the reported numbers seem fairly

large. How robust are these Some observations on the model The model is very stylized in several dimensions No maturity transformation (a key role of banks) Competition: the US banking industry is treated as a perfectly integrated one (no segmentation): is the Stackelberg-Cournot model the right one? Is welfare analysis feasible in this framework? Concluding remarks An ambitious paper with a long list of potentially useful extensions (Several papers, too much stuff in one only!)

The key issue is however: how much confidence can we place on the quantitative results? My suggestions: 1. Given a baseline model, provide a method to measure the errors associated with any estimate of policy counterfactual (at a minimum, extensive robustness analysis) 2. Harmed with such a method, richer versions of the model might allow comparisons of policy counterfactuals evaluated across nested models References

Begenau, Juliane, 2015, Capital Requirements, Risk Choice, and Liquidity Provision in a Business Cycle Model, Harvard Business School Working Paper, No. 15-072, March. Clerc, Laurent, Alexis Derviz, Caterina Mendicino, Stephane Moyen, Kalin Nikolov, Livio Stracca, Javier Suarez, Alexandros P. Vardoulakis, 2014, Capital Regulation in a Macroeconomic Model with Three Layers of Default, forthcoming in the International Journal of Central Banking De Nicol Gianni, Andrea Gamba, and Marcella Lucchetta, 2014, Microprudential Regulation in a dynamic model of banking, Review of Financial Studies, Vol. 27, 7, 2097-2138 Martinez-Miera, D. and J. Suarez (2014). Banks' endogenous systemic risk taking. Working paper. http://www.cemfi.es/~suarez/ Nguyen, Thien Tung, 2014, Capital Requirements: A Quantitative Analysis, mimeo,

Recently Viewed Presentations

  • Feb.26, 2019 You need: Clean paper / pencil

    Feb.26, 2019 You need: Clean paper / pencil

    Body Tube - gives distance between eyepiece and objective. 1. ... Analogy to City: City "Gate Keeper" ... Golgi Body. BOTH Plant and Animal Cells. Function: receives materials from the endoplasmic reticulum and sends them to other partsof the cell.
  • The Dimensions of Human Communication

    The Dimensions of Human Communication

    THE DIMENSIONS OF HUMAN COMMUNICATION B. Pragmatics** Practical use of language in social interaction Focuses on the speaker's achieving a practical outcome through using language as a tool Speaker and listener roles differ due to social context Justice & Redle...
  • Native Language Immersion Curriculum Development and Assessment

    Native Language Immersion Curriculum Development and Assessment

    Authorship and illustrator credentials clearly indicate if the materials were created by an American Indian in accordance with PL 101-164 of 1990 the Indian Arts and Crafts Act, and if this work was completed for, or on behalf of, an...
  • 1.12.15

    1.12.15

    What you need to knowSigmund Freud. List and describe the 3 components into which Freud divided the personailty & indicate how these are distributed across the 3 levels of awareness.. Explain the role of . internal conflicts . in Freud's...
  • Microsoft Azure: Infrastructure as a Service (IaaS) Module

    Microsoft Azure: Infrastructure as a Service (IaaS) Module

    Navigate to different portal areas, focusing on the dashboard. Next, go to the new Azure portal to show pre-existing resource groups or to create a resource group [EDITOR] TWB_Trevor: 04 October 2013. Content seems to be missing from the bullet...
  • Reviving Religion - DavidChapman.Org

    Reviving Religion - DavidChapman.Org

    The language of the declaration and the resolutions was too strong for most of the 300 in attendance. Only 1/3 (about 100) signed the document. Some of them will remove their names after a torrent of ridicule follows the convention....
  • What is a paragraph? - Skills Workshop

    What is a paragraph? - Skills Workshop

    What is a paragraph? A group of sentences about one main idea Includes a topic sentence that will indicate to the reader what the paragraph will be about Will have supporting sentences giving detail and evidence to support this idea...
  • Trapezoids and Kites - DavisEric.com

    Trapezoids and Kites - DavisEric.com

    Trapezoids Trapezoid Def: A Quadrilateral with exactly one pair of parallel sides. The parallel sides are called the bases. The non-parallel sides are called the legs. A trapezoid has two pairs of base angles. If the legs are congruent, then...