Why regulate the financial system?

By SHEILA DOW

This article was published in our Rethinking Finance publication, Check out the full magazine here

It may seem remarkable that we should still be asking whether the financial sector should have special regulation, so soon after the latest financial crisis and with the prospect of another one. Yet the debate continues, and indeed new post-crisis bank regulation is currently being rolled back in the US. So the argument for regulation still needs to be made. We will go back to first principles in order to consider why the financial sector needs special regulation, beyond normal company regulation (see further Dow (1), Kregel and Tonveronachi (2). How these principles are applied depends on the context, which, in finance, is always evolving. We will focus on the modern state of money and finance to consider issues surrounding the type of regulation that is needed.

WE MUST ALSO BEAR IN MIND the goals set for the monetary authorities when considering financial regulation. Before the crisis, the authorities focused primarily on targeting inflation. But since then, the authorities have been forced to prioritise dealing with the financial instability of the crisis and the economic stagnation which resulted from austerity fiscal policy. Monetary authorities still aim to promote monetary stability by controlling inflation. Additionally, they, formally and or informally, also target financial stability (e.g. the degree of volatility in asset prices, and financial conditions more generally) and economic stability (which is the primary goal set for Norway’s central bank).

“As far as finance is concerned it is argued on the basis of historical study that financial instability in the past has actually been caused by state interference”

HOW WE DISCUSS FINANCIAL REGULATION depends on how we understand the way in which the financial sector operates and how it fits in with general economic processes. We will explore in particular four different approaches which vary in the extent to which they see governments intervening in finance: the neo-Austrian approach, the mainstream New Keynesian approach, the Sovereign Money approach and the Post-Keynesian approach.

NEO-AUSTRIAN APPROACH

This approach advocates for as little government involvement in finance as possible, relying instead on market forces. The general approach is based on the idea that individuals in the market have better knowledge to guide their decisions than the state does. So according to Hayek, the role of the state is limited to ensuring that markets operate competitively:

“If man is not to do more harm than good in his efforts to improve the social order, he will have to learn that … where essential complexity of an organised kind prevails, he cannot acquire the full knowledge which would make mastery of the events possible. He will therefore have to use what knowledge he can achieve, not to shape the results as a craftsman shapes his handiwork, but rather to cultivate a growth by providing the appropriate environment, as the gardener does for his plants”(3)

NEO-AUSTRIANS see state supply of money as a source of monetary instability, inflation being seen as the byproduct of politically motivated increases in the money supply. They therefore argue for money to be supplied only by the banks in the form of deposits whose value would vary with the value of the banks’ assets. Competition would then determine which bank’s deposits were the most appealing to hold as money. There would no longer be any need for a central bank.

AS FAR AS FINANCE IS CONCERNED, it is argued on the basis of historical study that financial instability in the past has actually been caused by state interference. In terms of the recent crisis, the argument is that the state’s support of the banks created moral hazard: it encouraged banks to take undue risks. If instead banks were allowed to fail if they invested unwisely, their customers would withdraw their deposits when they sensed trouble ahead. In other words, the market would be a much more effective discipline on bank behaviour than any regulation (4).

NEW KEYNESIAN APPROACH

The New Keynesian approach is what we might call the current mainstream approach to financial regulation, in that it is the one which has dominated the theoretical literature and much of the policy debate. The previously dominant mainstream approaches (Monetarism, then New Classicism) had assumed, like the neo-Austrian approach, that efficient market forces would ensure financial stability. Indeed, because past and current mainstream approaches use models which focus on equilibrium, stability is built into them as the natural place for economies to settle. Also, like the neo-Austrian approach, mainstream approaches have continued to adopt a monetarist view of inflation: that it is caused by changes in the supply of money.

BUT THE New Keynesian approach argues that in practice markets do not work perfectly, so that regulation is needed to ensure financial stability. In particular, they challenge the view, both of neo-Austrians and of the previous mainstream, that individuals have perfect knowledge for making the best market decisions. One consequence is that banks may ration credit, in the absence of full knowledge about borrowers’ riskiness.

More significantly for them, in the run-up to the latest crisis, banks were unable to assess properly the riskiness of the complicated structured market products which ultimately proved to be highly risky. Critically for this approach, it is in principle possible to measure the riskiness of any asset, given full information. So, one aim of regulation is to enhance information availability, for example addressing the factors that lead credit-rating agencies to distort their ratings.

A FURTHER AIM OF REGULATION, as with the neo- Austrian approach, is to remove incentives to take on undue risk, notably the promise of central bank liquidity support. The New Keynesian approach to regulation has thus focused on dealing with bank failure. There has been a push, for example, for banks to issue contingent convertible (‘coco’) bonds which would in times of crisis convert debt into equity. This reflects the mainstream view that, because they are continually priced in competitive markets, equity is more efficient than debt (especially illiquid bank loans, but also bonds). Further along these lines, were banks to fail, they should be bailed in rather than bailed out, i.e. the risk of failure should be priced in to their liabilities. Such elements are seen to be desirable features of the resolution mechanisms (or ‘living wills’) advocated for banks, enhancing in turn the knowledge of risk on the part of depositors.

“Since monetary and financial instability are associated with credit cycles, the cure…is for the state to take over from the banks the supply of money.”

THERE HAS BEEN GROWING AWARENESS of network effects, whereby risks associated with one institution or asset can spread to others; these effects are a form of externality with respect to decision-making by any one institution. This is a further form of market imperfection for which New Keynesians have advocated regulatory reforms following the crisis. But Calomiris offers a critique of reforms which have been adopted, as departing from New Keynesian principles, the first of which, for him, is that:

“Financial regulation should focus exclusively on bona fide objectives that relate to the performance of the financial sector, grounded in core economic concepts of externalities and information costs and supported by evidence that shows that the costs of regulation are justified by demonstrable benefits” (5)

SOVEREIGN MONEY APPROACH

The third, Sovereign Money (or Positive Money), approach shares with the New Keynesian approach the view that bank behaviour, influenced by moral hazard, is the root cause of the financial crises (rather than the state, in the case of the neo-Austrian approach). Society’s money is mostly in the form of bank deposits, so that deposits are continuously recycled through the banks as payments are settled. This gives the banks the freedom to expand credit at will, with increased money the consequence. Rather than the money supply being controlled by the central bank, it is controlled by the banks: it is endogenous. Since monetary and financial instability are associated (as in the recent crisis) with credit cycles, the cure for both is for the state to take over from the banks the supply of money.

ADOPTING A MONETARY THEORY OF INFLATION, it is argued that state control of the money supply would allow direct control of inflation. Also, it is argued that the banks would no longer be able to cause financial instability. Having lost the power to create money, and the state support that historically has gone with that, banks would have to accept market discipline like other financial institutions. The Sovereign Money argument for the state to establish a monopoly over money would eliminate banking as we know it, without further need for regulation, or even deposit insurance6. Like for the neo-Austrians, bank liabilities would then vary in value along with the value of banks’ assets.

THIS PROPOSAL has much in common with proposals current in the 1930s (prompted similarly by a financial crisis) for what was called full reserve banking (7). Banks would be required to back their deposits hundred percent by reserves with the central bank. There were several variants of these proposals, but the basic principles were the same. In many cases, the vehicle for getting new money into the economy was via fiscal expenditure. In some cases, the system would allow for a negative rate of interest imposed by the state as a means of discouraging hoarding of money. In the case of Sovereign Money, the state would issue money rather than the banks, but the banks would administer it, and the payments system. However, the central bank would also coordinate with government by tying money supply to both fiscal policy and to allocating resources to facilitate and direct bank lending in pursuit of social and environmental goals, as well as economic goals. The functions of the central bank are thus envisaged to be much broader than has recently been the case.

POST-KEYNESIAN APPROACH

Post Keynesians share with the Sovereign Money approach this broader role for the monetary authorities, and also the view that the money supply is endogenously determined by the banks. But otherwise the analysis of the financial sector is very different (8).

SINCE the Post-Keynesian approach is the one which I advocate, I aim in what follows to explain why the other approaches are unsatisfactory from a Post-Keynesian point of view.

THE INTER-CONNECTEDNESS OF ECONOMIC SPHERES

THE POST-KEYNESIAN approach differs from the others, not just in terms of its theory of money and finance, and the policy proposals which follow, but also in terms of their basis in the way in which the economy is understood. In particular, the first three approaches involve conceptual separations: notably a separation between the state and the private sector, a separation between money and other assets, and a separation between money and finance on the one hand and the real economy on the other. Since Post Keynesians focus instead on interconnectedness, the three goals of monetary stability, financial stability and economic stability are seen as interconnected, and therefore all the business of the monetary authorities. Monetary stability is the least important, since a stable economy and stable financial conditions are the main ingredients of low inflation, the money supply being endogenous (9).

THE STATE AND THE PRIVATE SECTOR are intertwined, not least in terms of the market for sovereign debt and thus the implementation of monetary policy. More fundamentally, the state provides an institutional, legal and knowledge foundation for the private sector; indeed, the state has evolved along with markets to meet society’s needs. This is clear from the history of banking. As banking evolved, it became clear that banks operating at the micro, firm level, did not normally address the macro level. A central bank instead could see the macro consequences of banks’ actions and either attempt to moderate these actions or act to moderate the consequences. Thus, confidence in the system as a whole was provided by the central bank providing a lender-of-last- resort facility, so that the liquidity problems of one bank would not spread throughout the system, or indeed so that problems for the banking system as a whole would not cause a crisis. There was an implicit deal (a ‘social contract’) between the central bank and the banks that support would be guaranteed as long as banks accepted restrictive regulation and supervision to limit the need for support.

MANAGING UNCERTAINTY AND RISK

WHEN THIS SYSTEM worked well, the banks were able to supply society with a stable money asset in the form of bank deposits. Occasionally this kind of system could emerge without a state-run central bank, as in Scotland in the eighteenth century, when the older banks acted like a central bank towards the newer banks which threatened confidence in the system as a whole. Thus state-like institutions can evolve within the private sector if the state does not meet a need. The neo-Austrian approach focuses on financial markets at the micro level, such that any bank failure can be dealt with by transferring deposits to sound banks. If in fact there is scope for systemic bank failures, then market discipline is insufficient, and, without emergence of a private sector central bank, the system will collapse, leaving society without money.

“Money has the capacity both to enable real activity and to constrain it.”

THIS NEED FOR A SAFE MONEY asset is central to the Post Keynesian theory of money and banking. In contrast with the neo-Austrian and New Keynesian approaches, Post Keynesians emphasise the importance of the understanding that most of our knowledge is uncertain. This means that in general it is not possible to calculate the riskiness of any asset; rather than being concealed, as in the New Keynesian approach, such measures are unknowable. For Post Keynesians, it is unwarranted to rely on market pricing for financial stability under uncertainty.

FURTHER, where neo-Austrians and proponents of Sovereign Money envisage individuals making their own decisions about how to value their bank deposits, Post Keynesians argue that the ensuing uncertainty would make these deposits unsuitable as money. More generally, as Minsky (10) argued, the unavailability of true risk measures means that market valuations rely heavily on conventional judgements and are thus subject to wild swings. Since upward swings encourage increased leveraging (i.e. exposure to risk of asset price collapse), the outcome is financial instability as judgements about risks go into reverse and fire-sales of assets make price falls even worse. This was Minsky’s Financial Instability Hypothesis.

SOCIETY NEEDS A SAFE ASSET to hold in times of crisis, or even of increased uncertainty about the value of other assets. Money thus acts as a store of value, a necessary feature of a means of payment. It also acts as a unit of account, which provides a secure foundation for debt and labour contracts. But, while state-issued money normally performs these functions best, the financial sector is adept at providing near-moneys, particularly when the state attempts to control the supply of its own money.

The state can attempt to separate its money from rivals by requiring taxes to be paid in it. But the state cannot enforce a separation between its money and other assets, as presumed by the Sovereign Money approach.

Financial history demonstrates the ways in which the financial sector generates assets which are close money- substitutes, bank deposits being a notable case in point. But these near-monies are highly problematic if unregulated.

REAL ECONOMY EFFECTS

THE THIRD SEPARATION which Post Keynesians avoid is between money and finance on the one hand and the real economy on the other. In the New Keynesian approach in particular, the two only connect when there is a market imperfection, e.g. credit for real investment projects is rationed because of concealed information about risk. In the Post Keynesian approach, the interconnections are fundamental. Money has the capacity both to enable real activity and to constrain it. It enables by providing a safe asset as the basis for contracts and as a refuge from uncertainty. But by the same token it constrains by providing an alternative to positive expenditure decisions when uncertainty is high. Then effective demand is reduced, creating unemployment. In particular, Keynes argued that the short-termism of financial markets diverted finance from productive investment, with obvious real effects. Finally, the financial sector also has real effects when it promotes inequality of income and wealth.

FOR KEYNES, the aim of policy (e.g. on monetary reform) should be the efficient promotion of individual liberty and social justice: ‘The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes’ (11). Keynes thus saw central banking as operating within a wider remit than the narrow inflation-targeting version of monetary stability. Central banks rather were to work side-by-side with government in pursuing their goals. His approach indicates an enhanced role for regulation, given the financial sector’s capacity to provide near-money assets and its inherent tendency to be unstable and to promote a maldistribution of income and wealth.

THE POST KEYNESIAN APPROACH to regulation is not to separate the state from the private sector, the former possibly with a monopoly on money supply and the latter relied upon to promote social welfare through competitive markets. Rather it is to accept the intertwined nature of public sector and private sector banking, to promote a mutually supportive relationship between the two, to regulate particularly closely the provision of money-assets to ensure their safety, and to be alert to ongoing needs for new regulation as the financial environment evolves.

IN THE WAKE OF THE 2007-08 CRISIS

MANY OF THE CHANGES which have been discussed and even introduced in the wake of the crisis appear to follow these principles. New microprudential regulation aims to ensure more prudent bank behaviour, e.g. imposing minimum capital ratios, leverage ratios and liquidity ratios. But now there is also a much greater recognition of systemic risk which cannot be addressed at the level of individual banks. Macroprudential regulations have thus been introduced, including stress tests to identify individual banks’ vulnerability to adverse developments at the market level (12).

FUNCTIONAL SEPARATION between retail and investment banking is a further measure, designed to focus central bank support only on those institutions which supply society’s money. In fact, many of the other regulatory reforms too have been driven by the continuance of the ‘too-big-to-fail’ problem (that failure of a big bank would pose an unacceptable systemic risk). From a New Keynesian point of view, it is this major market imperfection which justifies continued regulatory intervention in the financial sector. There is now a greater appreciation in the mainstream policy literature of systemic risk, and of uncertainty. But the full significance of uncertainty for systemic risk which is central to the Post Keynesian approach to financial regulation is absent.

OUTSTANDING REGULATORY ISSUES FROM THE PERSPECTIVE OF POST-KEYNESIAN PRINCIPLES

One of the changes discussed but not implemented is a global tax on financial transactions, building on the original idea for a Tobin tax. As an added cost, such a tax would reduce the efficiency of the financial sector; such ‘sand in the wheels’ would be welcome from a Post- Keynesian perspective in that the sector would become less efficient in creating instability. Further, particularly if the revenues were used for redistributive purposes, the tax would increase the (Keynesian) efficiency to promote social justice. This is typical of the Post-Keynesian focus on efficiency with respect to social goals, rather than market efficiency, as the goal; it justifies introducing other constraints on markets, such as segmenting the financial sector through regulation, and capital controls, for example.

INCREASED CAPITAL REQUIREMENTS have been a continuing thread in the regulation discourse, relying on the mainstream notion of measurable risk of portfolios. There is debate as to the effectiveness of raising banks’ capital ratios. And it should not be forgotten that the recent crisis was fuelled by the strategies the banks had adopted in the face of increased capital requirements – securitisation of loans to get them off the balance sheet, diversion of attention, from lending to business, towards trading in securities and derivatives of various sorts. But what is perhaps most worrying is that blanket application of these minimum ratios to different types of financial institutions. In particular savings banks, co-operative banks and credit unions have struggled to meet new regulatory requirements more suited to large banks. And yet they were a notable source of stability throughout the crisis, providing support for the economic activities of their borrowers and depositors. These banks require to be segmented from the rest of the financial system regarding regulation – but also regarding the possibility of fiscal support.

IN THE MEANTIME, shadow banking which, by definition, is not covered by regulation, has expanded apace as a major source of credit, consequently increasing systemic risk. The clear implication is that regulation needs to keep up, and continue to keep up, to encompass shadow banking as it evolves. The Financial Stability Board is addressing the monitoring of shadow banking, which is a first step, but regulatory action is urgently required. Such regulation will be challenging, not least given the nimbleness of the financial sector, but that is not an adequate argument for no regulation.

DIGITAL CURRENCIES

The credit-creating development which has sparked most controversy recently regarding financial regulation is the emergence of digital currencies, or cryptocurrencies (see Weber for an excellent account) (13). While different payment vehicles, like credit cards and PayPal, had already offered alternatives to payment by cash or cheque, digital currencies also offer an alternative payments mechanism which is totally detached from the conventional central-bank-centred system. These new currencies are emerging at an accelerating rate as a means also of raising credit.

Are digital currencies examples of near moneys which threaten state control over money, and with it state earnings of seignorage?

IN FACT, they barely fulfill the three functions of money. Payment procedures are cumbersome (Bitcoin had to create a new ‘fork’, or spin-off, which would be less cumbersome for payments for goods and services), the value of each currency is highly unstable (even where the supply is fixed), and the unit-of-account function is impeded by the multiple valuations on different sites. Rather these currencies seem to be acting primarily as a vehicle for speculation. The proliferation of currencies, with new coin offerings (around 160 already in the first quarter of 2018) and forking from existing currencies, adds to the range of possible money assets, but also to the uncertainty surrounding their value. Further, these values tend to be correlated, creating a potentially fragile bubble within digital currencies.

INTERESTINGLY, there are attempts at pseudo-central bank regulation, such as the limit to Bitcoin supply and the way in which bitFlyer closes client’s positions when they lose half their original margin. But the prospect of this type of control extending to an ever-expanding array of currencies is hard to imagine, implying the need for regulation. In the meantime, derivatives markets have developed around currencies, exposing the overall structure to even more systemic risk.

“..digital currencies also offer an alternative payments mechanism which is totally detached from the conventional central-bank-centred system.”

THE PHILOSOPHY behind the first such currency, Bitcoin, is neo-Austrian. Society would provide its own money, independent of the state, founded on incentivised competition to validate transactions (the core feature of the new, distributed ledger, technology underpinning digital currencies). But the knowledge requirements for participants to value each currency is enormous. This is compounded by the need for individuals to be able to assess in each case the risk of fraud. This would all impose impossible demands on the general public, were digital currencies to be the only form of money.

CENTRAL BANK INVOLVEMENT

But what if the state were to issue their own digital currency (see e.g. Engert and Fung 2017)? (14) There has been detailed analysis by central banks of the merits of the distributed ledger technology used by digital currencies. But it is not clear how great the benefit would be, and indeed digital currencies are already relying on it less. But whether using this technology or not, central banks could establish digital currency accounts for the general public, through which payments would be routed, side- stepping the commercial banks. This idea holds much in common with the Sovereign Money proposal.

BUT THE STATE CANNOT ENFORCE a monopoly of money; inevitably other near-money assets would emerge to satisfy the need for a safe asset, including digital currencies, potentially earning a return. (The scope for a negative rate of interest is regarded as one of the attractions of a central bank digital currency.) The demand for such assets would rise when a rise in uncertainty provoked a rise in liquidity preference. If the supply of the central bank currency were restricted, or even just inflexible, it could not address discrete fluctuations in liquidity preference.

IF THE IDEA IS to remove credit-creating power from the banks by diverting money holdings to central bank accounts, the scope for financial instability would rise rather than fall. First, the central bank would no longer provide liquidity support in cases of deposit withdrawals, just as for other financial institutions at present.

Bank liabilities would then need to become marketable, such that they would only perform money functions poorly. And that would be the only recourse for increased liquidity preference other than the liabilities of even less regulated institutions offering attractive (yet potentially unsustainable) returns. Second, the demand for credit would also be diverted elsewhere fuelling the growth of shadow banking. The opportunity to direct bank credit to socially useful investment via new central bank currency would be welcome. But increased attention would need to be paid to the level and composition of credit creation elsewhere.

THE ALTERNATIVE Post-Keynesian policy of segmenting retail banking would take account of financial innovation leading to new providers of money assets and creators of credit. The aim therefore is to ensure that the payment and credit functions of banking within any institution are regulated in order to ensure the functionality of the system. Thus, for example, if fintech companies are performing these traditional banking functions, they should be regulated accordingly; their other functions need to be segmented off and made subject to regulation appropriate to those functions.

CONCLUSION

WE HAVE REVIEWED very different approaches to financial regulation. The three approaches other than the Post-Keynesian approach, all show excessive faith in the capacity of the market to ensure its own stability. Far too much reliance is placed on the price mechanism and the presumed ability to identify true measures of risk. The dominant, New Keynesian, approach sees the economy as inherently stable, but thrown off course by market imperfections. It is natural then to focus regulation on reducing those imperfections. The other two of the three differ primarily over whether money, as a separable asset, should be provided solely by the market, or solely by the state.

THE POST-KEYNESIAN VIEW is that each of these approaches is over-simplistic in assuming that conceptual separations used in their theoretical analysis are fully reflected in the real world. They ignore the fundamental interconnectedness between the state and the market, between assets arrayed along a liquidity spectrum and between money, finance and the real economy and society, and ignore the pervasive influence of fundamental uncertainty.

POST-KEYNESIAN REGULATORY PROPOSALS have included the reintroduction of the notion of institutional segmentation within the financial sector which was eroded from the 1970s, in particular between retail and investment banking. This would involve a return to the traditional deal between central banks and providers of money assets, whereby the former provides liquidity support to the latter in exchange for observing regulation designed to ensure the stable value of deposits.

Rather than focusing on the potential for bank failure, this approach focuses on positive measures to ensure successfully sound banking. For the state to take on a monopoly over the provision of money would instead prevent these newly stable banks from providing society’s money and the credit to finance real activity. Along these lines, there should also be regulatory segmentation between commercial banks and cooperative banks, given their very different culture and practices.

But more generally, concentrating on monetary reform, either in the form of no state money or only state money, distracts from the urgent attention which needs to be paid to regulating the rest of the financial sector. For Post Keynesians, the financial sector is inherently unstable; regulation can never eradicate that instability, but it can moderate it. While the other approaches focus on the high degree of leveraging banks currently enjoy, because their liabilities are money, they fail to address the extent of leverage elsewhere in the system; this leverage is all the more dangerous because of the multiple, opaque, uses made of the same collateral.

The answer then is two-fold. First regulation should be addressed at making sure that the financial sector can provide both a safe money asset and credit to finance real investment. Second it needs to be focused on widening the regulatory net to try to moderate the instability building up elsewhere in the system.

This article was published in our Rethinking Finance publication,
Check out the full magazine here

1. Dow, S C (1996) ‘Why the Banking System Should be Regulated’, Economic Journal, 104 (436): 698-707

2. Kregel, J and M Tonveronachi (2014) ‘Fundamental Principles of Financial Regulation and Supervision’, FESSUD Working Paper Series, no. 29.

3. Hayek, F A (1975) ‘Full Employment at any Price’, Hobart Paper, 45. London: IEA. (p.42)

4. Dowd, K (2009) Lessons from the Financial Crisis: A Libertarian Perspective. Libertarian Alliance, Online.

5. Calomiris, C W (2017) Reforming Financial Regulation after Dodd-Frank. New York: Manhatten Institute. (p.61)

6. Jackson, A and B Dyson (2012) Modernising Money. London: Positive Money.

7. Dow, S C (2016) ‘The Political Economy of Monetary Reform’, Cambridge Journal of Economics, 40 (5): 1363–76.

8. Fontana, G and M Sawyer (2016) ‘Full reserve banking: more “cranks” than “brave heretics”, Cambridge Journal of Economics, 40 (5): 1333–50.

9. Dow, S C (2017) ‘Central Banking in the 21th Century’, Cambridge Journal of Economics, 41 (6): 1539-57.

10. Minsky, H P (1986) Stabilizing an Unstable Economy. New Haven: Yale University Press.

11. Keynes, J M (1936) The General Theory of Employment, Interest and Money. London: Macmillan. (p.372)

12. Aikman, D, A G Haldane, M Hinterschweiger and S Kapadia (2018) ‘Rethinking financial stability’, Bank of England Staff Working Paper No. 712.

13. Weber, B (2016) ‘Bitcoin and the legitimacy crisis of money’. Cambridge Journal of Economics, 40: 17–41.

14. Engert, W and B Fung (2017) ‘Central Bank Digital Currency: Motivations and Implications’, Bank of Canada Staff Working Paper 2017-16.