Reflexivity, Financial Instability and Monetary Policy: A ‘Convention-Based’ Approach

ABSTRACT

The aim of this paper is to provide a theoretical analysis of the role of social conventions as emergent phenomena in financial markets, the latter being thought of as dynamically complex systems. Combining complexity and reflexivity with Keynes’s view of financial markets, we develop a ‘convention-based’ approach which shows how conventions can only temporarily stabilize the system, inevitably leading to financial instability and crises. Then, we adopt this framework to investigate a central banking agenda to act ‘against the tide’ (i.e., the prevailing convention) in the build-up of a speculative bubble, which calls for macroprudential policy and financial regulation. In this respect, we analyze the moral suasion channel that allows the central monetary authority to ‘talk down’ the market through asset-price management and the unconventional monetary policy toolkit.

KEYWORDS:

• Conventions
• complexity
• reflexivity
• financial instability
• monetary policy

JEL CODES:

• E12
• E44
• E58

Acknowledgements

We wish to thank Jan Toporowski, Duncan Foley, Anna Carabelli, Pietro Terna, Roberto Marchionatti, Mario Cedrini, and two anonymous referees for their insightful comments. We are also grateful to the remarks by the discussant and participants to the STOREP 16th Annual Conference held in Siena, June 27–29, where an earlier draft of the paper was presented. This work was supported by the Fondazione Luigi Einaudi Onlus under the Grant A theoretical and empirical analysis of complex dynamics in financial markets: a Post-Keynesian perspective. The usual disclaimer applies.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 Arthur (2014, p. 187) defines complexity economics as a theory wherein time becomes important and structures constantly form and re-form giving rise to a meso-layer between the micro- and the macro-structure, in which the emergence of unpredictable phenomena is considerably relevant due to network structures, feedback mechanisms and the heterogeneity of agents.

2 Recent studies in behavioral finance confirm that the expectation formation process is indeed extrapolative (see Gennaioli and Shleifer 2018).

3 Keynes (1936, p. 152, original emphasis) is perfectly clear on this when he claims ‘We are assuming, in effect, that the existing market valuation, however arrived at, is uniquely correct in relation to our existing knowledge of the facts which will influence the yield of the investment […]’.

4 The French School on conventions (Économie des conventions) arose as a branch of economic theory at the end of the 1980s, with pioneering contributions by Dupuy (1989), Eymard-Duvernay (1989), Favereau (1989) and Orléan (1989). See Dequech (2012) for a detailed account of this approach.

5 In a nutshell, the theory of reflexivity conceives subjective (individual) reality as connected to the objective reality (the state of affairs, directly observable) by means of two causal relationships: a ‘cognitive function’, whose causality goes from the world to the mind of the agents, and a ‘manipulative function’, which acts in the opposite direction. Rosser (2020) provides a detailed account of the theory of reflexivity and its relation to complexity.

6 As a matter of fact, taking a contrarian position to the prevailing one in the build-up of speculative bubble can result in sub-optimal monetary payoffs for investors. Accordingly, in presence of reputational sanctions, incentives to conform are even higher and a convention eventually becomes a social norm (see Keynes 1936, p. 158; Dequech 2011, pp. 485–487).

7 Self-referentiality is an instance of self-reflexivity, since a subjective aspect of reality feeds upon itself (Soros 2013, p. 313).

8 In accordance with Soros (2013), who defines a boom as a ‘far-from-equilibrium’ position.

9 A similar argument can be made with respect to the EMH, where what ultimately matters is not the existence of a fundamental in and of itself, but whether or not the market converges to it. The majority of empirical tests of the EMH are not able to reject the presence of serial correlations of stock returns at different holding periods, especially shorter horizons, confirming instead the presence of excess volatility and persistent price fluctuations.

10 Sau (2013) provides a detailed account of the Great Financial Crisis through the lenses of complex system theory and Minsky’s ‘Financial Instability Hypothesis’.

11 Soros (2013, p. 323) decomposes a speculative bubble into two components: ‘an underlying trend that prevails in reality’, and ‘a misconception relating to that trend’. When the feedback mechanism between these components ceases to be self-reinforcing, a boom turns into a bust.

12 This was the Federal Reserve’s declared policy prior to the subprime crisis (Greenspan 1999, 2002).

13 This view has been strongly advocated by the Bank of International Settlement (Cecchetti et al. 2000; Borio and Lowe 2002).

14 It is worth reminding ourselves that, for almost forty years, the mainstream approach to monetary policy has been dominated by the so-called ‘divine coincidence’ view (Blanchard 2016), according to which the manipulation of a single policy instrument, the overnight interbank interest rate, was supposed to accomplish three different goals: full employment, maximum sustainable growth and stable inflation. Neither asset-price stability nor a broad supervision of the financial system has ever been considered, let alone the implementation of discretionary policies, regarded as sources of disturbance and inefficiency as opposed to elements of strength. However, the 2008 crisis tilted the view toward how to react to asset price bubbles, thus clarifying that the central banking agenda cannot be restricted to an indirect promotion of fundamental valuation and deregulation and reduced to a unique goal, inflation targeting (see Everaert et al. 2019, p. 11).

15 We owe this insight into a referee.

16 The type of financial markets we have in mind are primarily those characterized by securitized stocks, where broad institutional investors play a crucial role and the investment horizon is sufficiently long. A representative example is the real estate market, whereby we can virtually conceive of some room for policy intervention in that the bearish/bullish convention thus generated can last for longer. On the contrary, if we think of speculation on overnight interest rates, it would be hard to imagine a feasible measure of intervention that the monetary authority could adopt to act upon such a short-lived convention.

17 As point out by Ciocca (2015), the emphasis on stability, subject to the constraint of not encouraging risky behaviour in the banking and financial industry, is already present in the first theoretical works on central banking by Henry Thornton (1802).

18 Broadly speaking, we should never forget the role of the central bank as a ‘lender of last resort’, which act by setting a floor to asset prices in order to attenuate a debt-deflation spiral. Indeed, if the central bank lends to a troubled financial institution, the latter does not have to sell its assets to meet creditors’ demands for redemption.

19 Different types of FG have been identified within the literature. Bernanke (2020, pp. 954–957) reviews their main features and discuss their effects on asset prices and government bond yields.

20 A moderate change in the interest rate reflect exactly Keynes’s (1936, pp. 203–204) recipe for a credible monetary policy.

21 In addition to reducing the volatility of the bond rates, this measure required a lower quantity of assets to be purchased to achieve the same rate reduction as compared to quantitative easing.

22 By contrast with mainstream economic theory, real investment, from a post-Keynesian perspective, is not driven by the interest rate for it mostly depends on the individual state of confidence, the ‘animal spirits’.

23 As a matter of fact, crises are most severe when accompanied by a lending boom and high leverage of market players, and when financial institutions themselves are participating in the buying frenzy.