Bankking System Regulated

8/6/2019 Bankking System Regulated

1/11

Th e Economic Journal, io6 [May), 698-707. Royal Economic Society 1996. Published by BlackwelPublishers, 108 Cowley Road, Oxford OX4 iJF, UK and 238 Main Street, Cambridge, MA 02142, USA.

WHY THE BANKING SYSTEM SHOULD BER E G U L A T E D *

Sheila C. DowI t has long been taken for granted that the financial system should beregulated. That this should now be open toquestion is to bewelcomed. It isimpor tant ingeneral to question assumptions, but it is particularly importantto do so in the case of a sector which has undergone such major change overthe last few decades. It is necessary therefore to go back to first principles andexamine the implications of these principles in the current environment. Theoutcome outlined here is a renewed case for regulation, but one which takesaccount ofthe changing nature and role of banks. Historical evidence aids ourun der stan din g of these changes, but historical evidence must be used w ith ca re;the advocates of free banking misinterpret and misuse this evidence to supporttheir position.

The case for regulation rests on the very special economic role of money andthe uncertainty associated with it. This uncertainty in turn renders freebanking unworkable since the proposal requires the non-bank public to assessthe expected value of the portfolios of the issuers of money. Adequateknowledge could only be generated if money-issuing were concentrated in adominant institution, or set of institutions, which operated like a central bank.This ou tcome w ould only occur, and be socially acceptab le, given a high degreeof social cohesion. R ath er tha n erad icating reg ulation on the groun ds th at it isflawed, and risking financial chaos, the more appropriate response is toconsider how to improve regulation.

I. BANKS AS PRODUCERS OF MONEYBecause banks' liabilities are used as money, their redeposit ratios aresufliciently high to allow them to hold longer-term assets without fear ofilliquidity. The primary purpose of prudential regulation has been to ensurethat banks' assets retain sufficient liquidity to meet any reduction in redeposits,and to discourage such a reduction in the first place. Regulation is warrantedbecause the moneyness of bank liabilities is a public good: moneyness (ratherthan any particular money asset) satisfies the conditions of non-rivalry-in-consumption and non-excludability-in-exchange. The state produces money-ness by inspiring confidence in money's capacity to retain value. Thisconfidence underpins the performance of money's role as means of payment


2/11

[ 19 96 ] W^HY T H E B A N K I N G SY ST E M SH O U L D B E R E G U L A T E D 69 9here in the sense of a learning gap which in principle can be closed. Rather itrefers to unq uantifiable risk - a lea rnin g g ap w hich can never, even inprinciple, be closed. Uncertainty in this sense pervades the economic process,colouring the decisions to consume, invest, supply labour and allocate wealth.In order for the economic process to proceed in spite of uncertainty, societyadopts conventions as a basis for expectation formation, and, supported by thestate, creates elements of stability to aid decision-making (see Hodgson, 1988).T he legal system suppo rts the establishm ent of con tracts, while the provision ofoutside money and bank regulation and supervision supports the evolution ofa banking system which produces money as an asset to hold in times ofparticular uncertainty. Just as money's role is integral to the economic process,so the state's role is integral to the evolution of private sector institutions andconventions. While it is conceivable that the private sector could evolve suchinstitutions and conventions without state support, and indeed can be said tohave done so in particular historical instances, there is no guarantee that itwould do so.Money is the ultimate object of liquidity preference. Understood in flow aswell as stock terms (see Dow, 1993, pp. 165-8), liquidity preference operatessimultaneously on both sides of the banks' balance sheet. In addition toinfluencing the demand for bank deposits, liquidity preference also influencesthe expenditure and borrowing plans of the non-bank sector, as well as thewillingness of banks to extend credit. A generalised increase in liquiditypreference thus creates a para do x of liquidity (see Dow , 1993, pp . 146 -52 ): jus twhen the demand for bank deposits rises because of increased uncertaintyabout asset values, their supply may be curtailed by a credit crumble. Thefinancial system thus exacerbates the downturn phase of the business cycle.Similarly, the reverse process exacerbates the expansion in output in upturns:banks are most willing to expand credit just when demand for holding liquidassets is at its lowest.The reserve ratio in many countries no longer acts as an eflective constrainton credit, due largely to the priority given to promoting confldence in theflnancial system through the lender-of-last-resort facility. Central banks canuse this facility to prevent an undue credit crumble when bank expectationsturn out to have been over-optimistic. The UK banking system has probablygone furthest down the road of reducing required cash ratios: it is now almostentirely an inside money system. The outsideness underpinning the bankingsystem is no longer a tangible metal, or even tangible liabilities of the publicsector, but the experience of bank regulation and supervision. The public'sconcern, insofar as it exists, is that deposits might not be accepted as a meansof payment because of concerns over the banks' actual or perceived solvency;a problem only of illiquidity, and perceived as such, can always be resolved byinterbank lending, or lending, by the central bank. That the UK public is not


3/11

700 THE ECONOMIC JOURNAL [ M A Yin a system which provides the vehicle, money, for the economic process, in asector where expectations of asset prices are crucial and yet not covered byquantifiable risk. The structure of the financial system has been undergoingmajor change, which certainly invites a major rethink about regulation. Butmoneyness itself is a convention, the product of centuries of experience. Topropose to remove these elements of stability and introduce a new conventionis bold to say the least; the confidence arising from many years' experience isnot to be dismissed lightly, nor can the behavioural effects of changingconventions be predicted easily.

II. C R I TI QU E OF F R EE BA N KI N G P R OP O S A LSThe proposal to free the banking system from regulation rests on a verydifferent view of mo ney from tha t outlined a bov e (see Sm ithin, 1994, ch ap ter2; Dow an d Sm ithin, 1994): one which presumes that un certa inty is eliminable.It follows first that there is no need for a perfectly hquid asset as a means ofpayment; bank liabihties of diversely varying value will do. Secondly, it followsthat financial assets can be valued in the same way as goods, so that there is noreason for banks to be regulated differently from other producers. Third, thereis no connection in their view between credit creation and the demand formoney (other than the increase in transactions demand which may follow froma credit-financed increase in expenditure), with no consequent risk ofinstability. Instability is therefore seen in micro terms as the result of unwiseportfolio decisions taken by individual banks, or imperfect information aboutindividual banks. While the law of refiux does not protect against a generalisedexpansion, Dowd (1993, pp. 212-7) argues that the risk of adverse clearingwithout a lender-of-last-resort would be enough to encourage individual banksto buck the trend. But it is the presumed absence of uncertainty which allowsDowd (1996) to argue that banks can actually detect overvaluation of assets(implying, for example, that banks in recent years knowingly entered into thesituation which generated bad debts).

Igno ring true un certain ty has pervasive impo rtanc e, not jus t for the n atu reof the demand for money and the propensity for systemic instability, but alsofor the operational feasibility of a free banking system. Free banking proposalsrest crucially on the market's capacity to value bank assets. In the absence ofstate regulation and supervision, it is the market which is to disciphne banksinto adopting prudent portfolios. Yet free bankers have not demonstrated thatthe market can actually generate the knowledge by which to assess bank risk;the w ord ' kno w ledge ' is used advisedly - it is not a m atter of com plete orincomplete information, but rather a matter of the abihty or otherwise topredict market values correctly, even within a probability distribution, and


4/11

199 6] VV HY TH E BA NK ING SYSTEM SHO ULD BE RE G U LA TE D 70 Ithat these valuations are contingent on a range of unknowns. There isaccordingly no sense in which one can talk in gen eral of the ' tr u e ' va lue of abank's assets. Keynes (1937) outlined the conventions employed in practiceunder uncertainty: giving undue weight to current trends, and relying onconventional judg em ents and on the judge m ents of experts. A conventionalview might emerge, through the pronouncements of pundits for example, thata bank's portfolio is unbalanced, with the effect of encouraging a run on thebank. But, as Davis (1992) points out, it is extraordinarily difficult even forcentral bank supervisors of banks (whose raison d'etre it is) to dete rm ine wh ethe ra ban k has a l iquidity problem or a solvency problem. T ha t is why prud entialregulation needs to be backed up by the lender-of-last-resort facility. G oo dh art(1993) argues in similar vein that private sector deposit insurers would faceinsurmountable difficulties in gathering adequate knowledge. It would be evenmore difficult for small deposit holders, and it is for this reason (withoutconsidering its wider implications) tha t Kaufm an (1991) advoca tes depositinsurance to reassure small depositors.

Recognising that unwarranted runs might occur, Dowd (1989, pp. 27-31)proposes, in the absence of an insolvency problem, that option clauses beinvoked when bank reserves are reduced. Banks would retain the option not toredeem their liabilities, with some compensation if the option were taken up.Dowd (1995) takes convertibility as an institutional given, necessary forstability at the level of individual banks and as a source of general pricestability. This would certainly be a reversal of recent b ank ing h istory, in whichthe notion of an outside medium of redemption has become progressively lessimportant. But if he is right that such a reversal is required, and if the publicare not assured of redemption on demand, it is not clear how much confidencewould be generated by the expectation th at redem ption would only normally beforthcoming. The use of the option clause, or even the possibility of its use,reduces the liquidity of ban k liabilities an d thus their moneyness. If redem ptionis as crucial for bank stability as Dowd argues, then option clauses wouldthreaten that stability.In order to satisfy the need for a safe asset, deposits would be attracted to theliabilities of those banks inspiring most confidence. While technical economiesof scale in banking are not sufficient to induce concentration, there aresignificant economies of scale in generating confidence. The conventionaljudgement on confidence-inspiring portfolios would tend to favour large, well-established ban ks. This jud ge m en t would be reinforced continually by itsconsequence of high redeposit ratios for these banks. The liabilities of thesebanks would then dominate as media of exchange, and would tend to be heldas reserves by the sma ller b ank s. Th e stability of the smaller bank s would then

depend on the stability of the large banks, and the capacity of the large banksto assess the credit-worthiness o fthe small banks and to discipline im pru den ce.


5/11

702 THE ECONOMIC JOURNAL [ M A Ythe banking sector with adequate liquidity, would address the issue of system-wide instability up to a point. But there would be no direct lending of last resortto individual banks, and no supervision; supervision is seen as being supersededby risk-assessment by (competing) federal insurance agencies. It would be upto the interbank market, then, to decide on the terms on which reserves mightbe borrowed. But it has not been demonstrated that the allocation of reservesby the market would be based on better knowledge than that available to acentral bank with a supervisory apparatus.

Already the banking system has moved in the direction of free banking byhaving capital adequacy ratios as the centrepiece of regulation. If banks wantto expand credit, then they must convince the market that it is warranted. Butthe m arket h as shown itself to make judg em ents wh ich, in agg regate, are atodds even with available, concrete information. The market was remarkablysluggish in absorbing even the most obvious implications ofthe debt crisis. Freebanking enthusiasts might argue that this myopia with respect to fundamentalsis induced by the sense of security provided by the state, i.e. by moral hazard.But the onus is on them to prove tha t, with out tha t sense of security, the ma rketwould have anticipated correctly the consequences of the increase in sovereigndebt in the 1970s, for example.Moral hazard is in any case far from pervasive in a regulated system. Banksfaced increased uncertainty in the 1970s and 1980s, first with increasedexchange instability, then with crumbling asset values, and finally withincreased competition due to deregulation. Even with deposit insurance andlender-of-last-resort facilities, banks expressed increased liquidity preference byshortening the term of assets, securitisation and so on (see Strange, 1986;Gardener, 1988). If such tactics were applied successfully (and that is a big' i f ' ) , they could protect banks from failure in a free banking system, but this isa far cry from Dowd's image of the prudent banking system meeting theeconomy's needs.Finally, free bankers argue that the risk of contagion does not justifyregulation. It is argued that deposits would simply be transferred from thebanks with unsound portfolios to banks with sounder portfolios, therebyencouraging sound portfolio management. But this argument ignores thepotential for systemic instability. As Minsky's (1982) theory elucidates, fragilityincreases for the financial system in general as the economy proceeds throughan expan sionary phase. Th us un realistic expectations of asset values on the p artof one bank are highly likely to be mirrored by unrealistic expectations on thepart of other banks. Any resulting contagion would not be unreasonable; itwould be the result of a change in conventional judgement about banks' assetvalues. Th is is wh ere the Be nsto n-K aufm an proposal is superior to the otherfree banking proposals: in the situation of a reversal in expectations about asset


6/11

1 9 9 6 ] W H Y T H E B A N K I N G S YS TE M S H O U L D B E R E G U L A T E D 7 0 3III . THE HISTORICAL EVIDENCE

The historical evidence both of regulated and unregulated banking systemslends support to these arguments, contrary to the claims of the free bankers.However, historical evidence should only be addressed with great care tothought experiments designed for modern application. Chick (1992; 1993)argues convincingly that banking systems proceed through an evolutionaryprocess, du ring w hich the n atu re of causal forces chang es. Th e need to gene rateconfidence in banking systems has encouraged the state to evolve a regulatorypresence, but the outcome has been an ever-increasing capacity in the privatesector to determine the volume and allocation of credit. This capacity iscompounded by the massive financial superstructure built on the foundation ofbank deposits as ultimate liquidity. The scope for systemic instability in theabsence of countervailing action by the state is accordingly increased. Further,the notion of commodities, or commodity-linked assets, as bank reserves hasbeen replaced by a com bination of gov ernm ent liabilities and confidence in thestate's capacity to encourage prudent bank behaviour. The free bankingsuppo rters are thus proposing to ju m p b ack from Chick's seventh stage ofbanking development to stage three.Few would de ny tha t there h ave be en periods of significant financialinstability when banks have been regulated (see Minsky, 1982; Kindleberger,1978; Davis, 1992). The case is easiest to make with respect to the UnitedStates, where the regulatory system has been conditioned by a dual state-federal structure and strongly conflicting interests. Yet it does not follow at alltha t the only reason able a lternative is no regulation . M an y episodes of financialinstability can be traced to misguided attempts to use regulatory power tocontrol the money supply; since such a project is unworkable (see Goodhart,1994), it is not surprising that it should produce instability. Yet, free bankersshare the monetarist view (applicable if at all only to early stages of bankingdeve lopm ent) tha t infiation is determ ined by the money supply which in turnis determined by bank reserves. Money supply regulation is accordingly thefocus of their attentio n (see Benston, 1991, and K aufm an, 1991). T he im portof the argument presented here is that more attention needs to be paid insteadto prudential regulation.Benston and Kaufman (1995, 1996) conclude from the US evidence thatfinancial fragility is not the result of bank behaviour since it has tended tofollow rather than precede failures in the real economy. But their evidence offinancial instability accentuating instability in the real sector, far fromcontradicting Minsky's theory, actually supports it. It is in the nature ofsystemic fragility that the real and financial cycles are closely interrelated.Cred it expansion to finance produ ctive activity is diverted to unp rodu ctiveactivity when the real cycle turns around; financial euphoria takes on a life of


7/11

704 THE ECONOMIC JOURNAL [MAYbubble can be identified with the failure of individual financial institutions, itis systemic fragility wh ich acc oun ts for the systemic consequences of such microfailures.The evidence of historical episodes of free banking is perhaps most pertinentto the debate . There has been some dispute in the l i terature as to what actuallyconstitutes an episode of free bank ing . For example, the experience of Scotlandin 1695-1845 cannot be isolated from the presence of the Bank of England.The Canadian free banking experience of 1820-1935 was coloured by the factthe Bank of Upper Canada and the Bank of Montreal were owned to asignificant extent by politicians in Upper Canada and Lower Canadarespectively.

At issue here is not just the question of how to interpret evidence, but thedeep er question of w hethe r the puris t version of free b an kin g as free co m pe titionis actually meaningful. Conventions and regulation exist along a cont inuum,with the only ultimate distinction being that it is the state which enactsregulat ion; but then the state and centres of power in the private sector alsoexist along a cont inuum. It is not legitimate to focus, as Dowd (1996) does, ona dual of regulation/laissez-faire. Given the central historical role of the stateand of long-established conventions in the financial sector, the onus is on freebankers to demonstrate the merits of overturning history, not the reverse.Indeed the successful episodes of free banking were those which evolvedconventions which closely mirrored state regulation. There has been a widerange of free banking episodes, but the two to receive most attention untilrecently are the United States case and the Scottish case. There is dispute as towhether the United States experience constituted a success (see for exampleGorton, 1985; Dowd, 1989, chapter 5), but the Scottish experience has beenwidely used to support influential arguments for free banking (see for exampleWhi te , 1984; Dowd, 1989; Hayek, 1990).While the Scottish case can indeed be interpreted as successful, it can also beseen to illustrate the arguments made above about the unworkabili ty of thetype of free banking scenario put forward by Dowd (1996). (See Dow andSmithin, 1992.) First, the banking system evolved its own form of centralbanking. The two 'old banks ' , the Bank of Scotland and the Royal Bank ofScotland, set up a note exchange and an exchange equalisation system,accepted the notes of other banks to curtail runs (most notably in the case ofth e Ayr Bank), disciplined banks whose credit creation was judged excessive(by refusing access to clearing) and made representations to Westminster onbehalf of the banking system as a whole. The other banks in turn held theirreserves with the old banks, which adopted an authoritative role, based onm arket pow er, which does not square with Dow d's (1996) ' cl ub ' interp retation.While this system succeeded because central banking evolved naturally in


8/11

1 9 9 6 ] W H Y THEB A N K I N G SY ST E M SH O U L D BE R E G U L A T E D 7O5by a profound sense of com m unity and the poten tial for strong social sanctions.But itwould be hard to argue that this would be a good general description ofthe context of free banking were it to be introduced now. Without an elementof public-spiritedness am ong leading bank s, it would be quite possible for m oralhazard to dominate , so that central banking and thus scope for financialstability would not naturally emerge.The Scottish experience also illustrates two further points. First, Dowd(1989) lays muc h stress on the use of option clauses to preem pt ban k ru ns. T heclauses was used widely in Scotland inthe early 1760s by private banks whichwere expanding credit rapidly. The 1765 legislation forbidding use of optionclauses was enacted because representations were made to the old banks,following public meetings, to lobby Westminster; the uncertainty attached tothe notes' value made them unacceptable to the general public. Indeed evenafter 1765 pub lic meetings resulted in boycotts of notes othe r th an those of theBank of Scotland and the Royal Bank of Scotland because of continuingunc ertainty as to their value. Option clauses and unce rtain value had removedthe moneyness of the private banks' notes.Secondly, the consequence of the old banks' notes being the standard ofvalue encouraged concentration of the banking system. Certainly there wereperiods of competit ion, but each wave of entry was followed by a process ofconcentration. Such concentration is the natural consequence of the need tosatisfy the demand for a stable money asset. It is evident in a wide range ofbanking histories, and is currently evident in the United States and inEuropefollowing the first fiush of com petition enco uraged by dereg ulation. T he re is noreason to doubt that a free banking experiment would also end up inconcentration in the banking system.Finally, inmore recent times, the debt crisis illustrates further the need forcentral banking functions which naturally arises in a deregulated environment.Sovereign debt expanded in the 1970s on a fiimsy knowledge base. Whenbanks' asset values collapsed, they were faced with the dilemma of whether towrite debts off or to reschedule, and concerns were voiced about the stabilityof the international financial system. The solution was found by makingrecourse to the IMF, whose knowledge base is much more extensive than thatof individual banks, which can take a system-wide view, and which has awell-established supervisory apparatus. Faced with a systemic problem, the marketdemonstrated its need for a world central bank which even goes beyond theremit of national central banks.Thus , at best, a totally deregulated market would throw up its own versionof a cen tral b ank , w ith its provision of a stan da rd of value a nd its supervisory-regulatory presence. But the long struggle in the United States to establish theFed eral R eserve System p rovides just on e exam ple of imp edim ents in the wayof the emergence of central banking.


9/11

7O6 T H E E C O N O M IC J O U R N A L [ M A Yprovision of lender-of-last-resort facilities to the system as a whole. It is anargument for regulation on the grounds that the moneyness which is necessaryto the working of the monetary system is a public good. However it is not anargument for complacency about bank regulation; regulation by its nature isproblematic and certainly involves costs (see Gowland, 1990). But economistsshould feel comfortable with addressing trade-offs between costs and benefits,and suspicious of any proposal which is presented as a cost-less solution.T he reg ulato ry issues facing us now arise from m ark et stru ctu ral diffusion(see Gardener, 1988), itself the product partly of deregulation. This diffusionhas altered the place of banks in the financial system and changed theknowledge requirements of the ma rket a nd of bank regulators. Experience w ithattempts to promote European integration in banking demonstrates amply thesignificance of different na tiona l m arke t conven tions, which in tu rn arecomplicated by different regulatory traditions. What generates confidence inone tradition is not necessarily what generates confidence in another tradition.Th ere is the additional question re garding the insti tutional se paration of banksupervision and attendant regulation from efforts to control monetaryaggregates.

Urgent regulatory issues are therefore posed by the potential for financialinstability which has arisen with a partial dismantling of regulation and fromthe changing role and behaviour of banks themselves. To suggest a completeremoval of regulation as the solution to these issues is both to admit defeat andto court disaster.University of Stirling

REFERENCESBenston, G.J. (1991). 'Does bank regulation produce stability? Lessons from the United States.' InUnregulated Banking: Chaos or Order'? (ed. F. Capie and G. E. Wood.) New York: St Martin 's PressBenston, G. J. and K aufm an, G. G. (1995). ' Is the paym ents system fragile ?' Em ory and Loy ola Universities(mimeo).Benston, G.J. and Kaufman, G. G. (1996). 'The appropriate role of bank regulation. ' ECONOMIC JOURNAL,this issue, pp. 688-697.Chiek, V. (1992). 'T h e evolution of the bank ing system and the theory of saving, investment an d interest.In Money, Method and Keynes: Selected Essays of Victoria Chick, (ed. P. Arestis and S. C. Dow). LonMacmillan.Chick, V. (1993). 'The evolution of the banking system and the theory of monetary policy.' In MonetaryTheory an d Monetary Policy: N ew Tracks for the iggos. (ed. S. F. Frowen). London: Macmillan.Davidson, P. (1978). Money and the Real World. London: Macmillan.Davis, E. P. (1992). Debt, Financial Fragility an d Systemic Risk. Oxford: Clarendon.Dow, S. C. (1993). Money and the Economic Process. Aldershot: Elgar.Dow, S. C. and Sm ithin, J. (1992). 'Free ban king in Scotland 1695-1845.' Scottish Journal of Political Econom

vol. 39 (November), pp. 374-91.Dow, S. C. and Sm ithin, J. (1994). 'Ch ang e in financial m arkets and the "first p rinciples " of mon etaryeconomics.' Universities of Stirling and York (mimeo).Dowd, K. (1989). The State and the Monetary System. Hemel Hempstead: Philip Allan.Dowd, K. (1993). Laissez-Faire Banking. London: Routledge.


10/11

1996] WHY THE BANKING SYSTEM SHOULD BE REGULATED 707Gorton, G. (1985). 'Banking theory and free banking history: a review essay.' Journal of Monetary Economics,vol. 16, pp. 267-76.Gowland, D. (1990). Th e Regulation of Financial Markets in the iggos. Aldershot: Elgar.von Hayek , F. A. (1990). Denationalisation of Money - Th e Argument Refined. London: IEA.Heinsohn, G. and Steiger, O. (1989). 'The veil of bar te r : the solution to "the task of obtaining

representations of an economy inwhich money is essential". ' In Infiation an d Distribution in CapitalistCrisis: Essays in Memory of Sidney Weintraub (ed. J. A . Kre gel). New Y ork: N ew York U niversity Press.Hodgson, G. M. (1988). Economics and Institutions. Oxford: Polity.Ka ufm an, G. G. (1991). 'Len de r of last resort incontemporary perspective. ' Journal of Financial ServicesResearch, vol. 5 (October) , pp. 95-110.Keynes, J. M. (1937). 'Th e Gen eral Theory of Em ploym ent. ' Quarterly Journal of Economics, vol. 51, pp.209-23.Kin dlebe rger, G. P. (1978). Manias, Panics and Crashes. New York: Basic Books.Minsky, H. P. (1982). Infiation, Recession an d Economic Policy. Brighton: WheatsheafSmithin, J. (1994). Controversies in Monetary Economics: Ideas, Issues and Policy. Aldershot: Elgar.Strange, S. (1986). Casino Capitalism. Oxford: Blackwell.White, L. H. (1984). Eree Banking in Britain: Theory, Experience and Debate, 1800-1845. Cambridge: CambridgeUniversity Press.


11/11

Documents

Bankking System Regulated