Banks, market organization, and macroeconomic performance: An agent-based computational analysis

https://doi.org/10.1016/j.jebo.2016.12.023Get rights and content

Highlights

  • An agent-based macroeconomic model with a banking sector is developed, calibrated, and simulated.

  • Banks improve macroeconomic performance by financing both new and existing trading firms.

  • The model normally tracks full-capacity utilization, but it occasionally exhibits systemic breakdowns.

  • In these worst-case scenarios, the economy performs better when banks are lightly regulated.

Abstract

This paper is an exploratory analysis of the role that banks play in supporting what Jevons called the “mechanism of exchange.” It considers a model economy in which exchange activities are facilitated and coordinated by a self-organizing network of entrepreneurial trading firms. Collectively, these firms play the part of the Walrasian auctioneer, matching buyers with sellers and helping the economy to reach prices at which peoples’ trading plans are mutually compatible. Banks affect macroeconomic performance in this economy because their lending activities facilitate the entry and influence the exit decisions of trading firms. Both entry and exit have ambiguous effects on performance, and we resort to computational analysis to understand how they are resolved. Our analysis draws an important distinction between normal and worst-case scenarios, with the economy experiencing systemic breakdowns in the latter. We show that banks can provide a “financial stabilizer” that more than counteracts the familiar financial accelerator, and that the stabilizing role of the banking system is particularly apparent in worst-case scenarios. In line with this result, we also find that under less restrictive lending standards banks are able to more effectively improve macroeconomic performance in the worst-case scenarios.

Introduction

How do banks affect the macroeconomy? The vast literature on finance and macroeconomics has provided numerous answers to this question, many of which are common knowledge. This paper develops an agent-based computational model to examine a relatively unexplored channel through which banks can influence macroeconomic performance, namely, their role in what Jevons called the “mechanism of exchange.”

In any but the most primitive economic system, exchange activities are organized by a network of specialized enterprises, not just the firms that produce goods and services but also retailers, wholesalers, brokers, and various other intermediaries. These enterprises provide facilities for trading at publicly known times and places, provide implicit guarantees of quality and availability of spare parts and advice, quote and advertise prices, and hold inventories that provide some assurance to others that they can buy at times of their own choosing. In short, they play the role in real time that general equilibrium theory assumes is played in meta time by “the auctioneer,” namely that of matching buyers with sellers and establishing prices that coordinate peoples’ trading plans. Moreover, unlike the auctioneer, these intermediaries provide facilities and buffer stocks that allow trading to proceed even when individual plans are imperfectly aligned.

The importance of this network of trading enterprises is attested to by Wallis and North (1986), who show that providing transaction services is the major activity of business firms in the U.S. economy; they estimate that over half of measured GDP in the U.S. consists of resources used up by the transaction process. Indeed, as everyday experience of any household will verify, almost all transactions in a modern economy involve enterprises that specialize in making these transactions.

Banks play a critical role in an economy's trading network, not just because they themselves are part of the network, intermediating between surplus and deficit units, but also because their lending activities influence the entry and exit of other enterprises that make up the network. Entry of new facilities is neither free nor automatic. It requires entrepreneurship, which is not available in unlimited supply and which frequently needs finance. Likewise, exit of existing facilities constitutes a loss of organizational capital that affects the system's performance, and exit is often triggered by banks deciding when to cut off finance from a failing enterprise.

The present paper describes a model that portrays this role of banks in helping to coordinate the economy. In a sense, our work is a continuation of a line of research into disequilibrium macroeconomics that began with Patinkin (1956, ch. 13) and Clower (1965), and reached its pinnacle in the Barro–Grossman (1976) book. That line of research ran into the problem that the failure of one market to clear generates rationing constraints that affect traders in other markets in complicated ways that are analytically intractable. To deal with such complexities, we have chosen to model the mechanism of exchange in an agent-based computational framework.1

More specifically, we use a modified version of the model originally developed by Howitt and Clower (2000), in which an economy's network of trade specialists was shown to be self-organizing and self-regulating. Howitt and Clower find that starting from an initial situation in which there is no trading network, such a network will often emerge endogenously and will, in the absence of shocks, also guide the economy to a stationary state in which almost all the gains from trade are fully realized. Here, we extend the model to allow for durable goods, fiat money, and government bonds, to include monetary and fiscal authorities, and to incorporate banks that lend to the trade specialists.2 We additionally introduce various random shocks that prevent the system from ever settling into anything like Howitt and Clower's fully coordinated stationary state. We calibrate the model to U.S. data and simulate it many times under different parameter values to see how banks and various dimensions of their lending behavior affect macroeconomic performance.

Although our model is admittedly too stylized to be used for policy-making purposes, it does produce three interesting results. First, it provides a framework for understanding “rare disasters.” Most of the time the evolving network of trade intermediaries performs reasonably well in counteracting shocks and keeping the economy in the neighborhood of full capacity utilization, but in a small fraction of simulation runs, the economy spirals out of control. The model thus exhibits what Leijonhufvud (1973) called “corridor effects” – that is, the system's self-regulating mechanism is unable to counteract shocks beyond a certain point. The distinction between normal and worst-case scenarios shows up dramatically in almost all the experiments we perform on the model.

Our second result concerns the stabilizing influence of banks on macroeconomic performance, especially in worst-case scenarios. It is generally accepted that although finance may help promote economic growth and development, this long-run benefit comes at a cost of increased short-run volatility. This notion is embodied in the basic idea of the financial accelerator that Williamson, 1987, Bernanke and Gertler, 1989, Holmstrom and Tirole, 1997, Kiyotaki and Moore, 1997, and others have shown can amplify the effects of macroeconomic shocks because of the endogenous nature of collateral. Our model, however, shows that banks that make collateralized loans to business firms also provide an important “financial stabilizer,” which can potentially be more powerful than the financial accelerator. In particular, when a negative macroeconomic shock is accompanied by firm failures, the presence of banks that can finance replacement firms and sustain other existing firms will often dampen the effects of the shock by ameliorating or even averting a secondary wave of failures that would amplify the drop in output and employment.

Finally, related to the role of banks as financial stabilizers, our third result is that under less restrictive lending standards, due to either higher loan-to-value ratios or lower capital requirements, banks are able to more effectively improve macroeconomic performance in the worst-case scenarios. Thus, in bad states of the world, there exists a conflict between micro-prudential bank regulation and macroeconomic stability.

The next section contains a brief literature review. Section 3 discusses the basic elements of our model. Section 4 describes the protocol by which agents interact in the model, as well as the behavioral rules that we are imputing to them. Section 5 describes a full capacity utilization stationary state that the system approximates in the hypothetical absence of shocks, and it also discusses the ways in which entry, exit, and bank lending affect the economy's performance. Section 6 describes how we calibrate our model and discusses its ability to match various empirical facts. Section 7 reveals our main results. Section 8 considers the robustness of our results to alternative assumptions regarding the banking system, and Section 9 concludes. Additional modeling details and supplementary results, including a sensitivity analysis, are relegated to appendix sections.

Section snippets

Previous literature

There is a large literature on the effects of financial intermediation on long-term growth.3 Our paper focuses on fluctuations rather than growth, and it therefore speaks more to the empirical literature on the effects of financial development on stability (e.g., Easterly et al., 2001, Braun and Larrain, 2005, Raddatz, 2006, Loayza and Raddatz, 2007). In particular, our notion of banks as financial

The model

Our model portrays, in an admittedly crude form, the “mechanism of exchange” through which economic activities are conducted in a decentralized economy. At the heart of the model is a self-organizing network of firms that coordinate all production and trading activities. Macroeconomic fluctuations in the model arise, in part, from disruptions to this network due to firm turnover, resulting in the break-up of established trading relationships in both labor and goods markets. Since exit and entry

Protocol and behavioral rules

Every week, the actors in our model proceed sequentially through the following nine stages: (1) entry, (2) search and matching, (3) financial market trading, (4) labor and goods market trading, (5) monetary policy, (6) match breakups, (7) fiscal policy, (8) exit, (9) wage and price setting. The rest of this section describes the protocol and behavioral rules governing agents’ actions at each stage.

A full-capacity stationary state

In our model economy, GDP is the sum of all goods produced.33 As shown in the similar model of Howitt and Clower (2000), GDP will be maximized when there is exactly one shop trading each good (which economizes on fixed costs) and everyone either has an employer or is one. In this case, GDP will equal full-capacity output Y*, which is total labor input minus the

Calibration and empirical validation

This section first describes how we calibrate our model and then discusses its ability to match various stylized facts observed in real-world macroeconomic data.34

The main results

Having empirically validated our calibrated model, we next perform our main experiments. We conduct a total of 10,000 simulation runs, each allowed to continue for T = 2880 weeks (60 years). All runs are initiated near the hypothetical stationary state of the economy described in Section 5.1. We do not start tabulating results from any run until 20 years have passed, in order to provide the system with sufficient time to possibly settle into a stochastic steady state. Fig. 2 shows the time series

Alternative assumptions regarding bank behavior

In Appendix E, we show that our main results with respect to macroeconomic performance under safe and risky banks in normal and worst-case scenarios are robust to 25 percent variations in almost all of the model's parameters. The present section demonstrates that our main results are also robust to alternative, and arguably more realistic, assumptions regarding bank behavior.

Conclusion

In this paper, we developed an agent-based macroeconomic model, focusing on the mechanism that normally makes a free-market economic system self-organizing and self-regulating, namely, its network of specialist trading enterprises. We applied our model to explore the role that the banking system plays in supporting this mechanism.

Our investigation generates a number of interesting results. First, it provides a framework for understanding rare disasters: states of the world in which the market

References (80)

  • R. Fonseca et al.

    Entrepreneurship, start-up costs and employment

    Eur. Econ. Rev.

    (2001)
  • M. Gertler et al.

    Financial intermediation and credit policy in business cycle analysis

  • P. Howitt et al.

    The emergence of economic organization

    J. Econ. Behav. Organ.

    (2000)
  • A. Leijonhufvud

    Agent-based macro

  • R. Levine

    Finance and growth: theory and evidence

  • S.C. Myers et al.

    Corporate financing and investment decisions when firms have information that investors do not have

    J. Financ. Econ.

    (1984)
  • C. Raddatz

    Liquidity needs and vulnerability to financial underdevelopment

    J. Financ. Econ.

    (2006)
  • V.A. Ramey et al.

    Inventories

  • J.J. Rotemberg et al.

    The cyclical behavior of prices and costs

  • L. Shyam-Sunder et al.

    Testing static tradeoff against pecking order models of capital structure

    J. Financ. Econ.

    (1999)
  • J.H. Stock et al.

    Business cycle fluctuations in U.S. macroeconomic time series

  • J.B. Taylor

    Discretion versus policy rules in practice

    Carnegie-Rochester Conf. Ser. Public Policy

    (1993)
  • J. Wiklund et al.

    The age-effect of financial indicators as buffers against the liability of newness

    J. Bus. Ventur.

    (2010)
  • P. Aghion et al.

    The Economics of Growth

    (2009)
  • G.A. Akerlof et al.

    Fairness and unemployment

    Am. Econ. Rev.: Pap. Proc.

    (1988)
  • E.I. Altman

    Financial ratios, discriminant analysis and the prediction of corporate bankruptcy

    J. Finance

    (1968)
  • Q. Ashraf et al.

    How inflation affects macroeconomic performance: an agent-based computational investigation

    Macroecon. Dyn.

    (2016)
  • J.S. Austin et al.

    The determinants of entry and exit rates into U.S. manufacturing industries

    Rev. Ind. Organ.

    (1990)
  • Bank for International Settlements

    Capital Requirements and Bank Behaviour: The Impact of the Basle Accord

    (1999)
  • R.J. Barro et al.

    Money, Employment, and Inflation

    (1976)
  • N. Basu et al.

    ASPEN: a microsimulation model of the economy

    Comput. Econ.

    (1998)
  • W.J. Baumol

    The Microtheory of Innovative Entrepreneurship

    (2010)
  • W.H. Beaver et al.

    Have financial statements become less informative? Evidence from the ability of financial ratios to predict bankruptcy

    Rev. Account. Stud.

    (2005)
  • R.L. Bennett et al.

    A Microsimulated Transactions Model of the United States Economy

    (1986)
  • B. Bernanke et al.

    Agency costs, net worth, and business fluctuations

    Am. Econ. Rev.

    (1989)
  • M. Bils

    The cyclical behavior of marginal cost and price

    Am. Econ. Rev.

    (1987)
  • M. Bils et al.

    Some evidence on the importance of sticky prices

    J. Polit. Econ.

    (2004)
  • H. Bohn

    The behavior of U.S. public debt and deficits

    Q. J. Econ.

    (1998)
  • M. Braun et al.

    Finance and the business cycle: international, inter-industry evidence

    J. Finance

    (2005)
  • A. Bris et al.

    The costs of bankruptcy: Chapter 7 liquidation versus Chapter 11 reorganization

    J. Finance

    (2006)
  • Cited by (0)

    We acknowledge valuable suggestions from the editors, Nick Vriend and Mauro Napoletano, and several anonymous referees, in addition to those provided by Blake LeBaron, Ned Phelps, Bob Tetlow, and various seminar participants at the Federal Reserve Bank of Cleveland, the Marschak Colloquium at UCLA, the Center on Capitalism and Society at Columbia, Brandeis University, Columbia Business School, the Federal Reserve Bank of Dallas, the National Bank of Austria, Brown, and MIT, none of whom bears any responsibility for errors or should be taken to endorse our analysis. The C++ code for our computational model is available online at https://drive.google.com/file/d/0Bw0xeVVYEG5XRHdNb19fYUkxYlE/view?usp=sharing. It was compiled and run as a 64-bit Windows application.

    View full text