Asset price bubbles and counter-cyclical monetary policy: Why central banks have been wrong and what should be done

Palley, Thomas I. Article Asset price bubbles and counter-cyclical monetary policy: Why central banks have been wrong and what should be done Intervention. European Journal of Economics and Economic Policies Provided in Cooperation with: Edward Elgar Publishing Suggested Citation: Palley, Thomas I. (2010) : Asset price bubbles and counter-cyclical monetary policy: Why central banks have been wrong and what should be done, Intervention. European Journal of Economics and Economic Policies, ISSN 2195-3376, Metropolis-Verlag, Marburg, Vol. 07, Iss. 1, pp. 91-107, https://doi.org/10.4337/ejeep.2010.01.09 This Version is available at: https://hdl.handle.net/10419/277178 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Asset price bubbles and counter-cyclical monetary policy: Why central banks have been wrong and what should be done Th omas I. Palley* Central banks have generally opposed targeting asset and credit market excess. Th is paper argues against that position. Bubbles can impose signifi cant harm through the debt footprint eff ects they leave behind, and through distortions resulting from using interest rates to mitigate their aggregate demand impacts. Conventional interest rate policy is not well suited to managing bubbles, and the paper argues for adoption of a new system of asset based reserve requirements (ABRR). Not only can ABRR target asset market excess, they also strengthen counter-cyclical monetary policy. JEL classifi cations: E52, E58, E60 Keywords: asset price bubbles, asset based reserve requirements 1. Reconstructing monetary policy after the Great Recession For the last several years central bank thinking has been dominated by infl ation targeting. Th e US, which was ground-zero for the fi nancial crisis, made infl ation its primary focus even though it stopped short of a formal infl ation target. Side-by-side with this focus on infl ation there was explicit opposition to targeting asset markets and asset price bubbles from both * New America Foundation, Washington DC. Th e author thanks two anonymous referees for comments that have greatly improved the paper. Responsibility for any errors is the author’s. Correspondence Address: Th omas I. Palley, 1913 S Street NW, Washington, DC 20009, USA, e-mail: [email protected]. Received 04 May 2009, accepted 02 September 2009 © INTERVENTION 7 (1), 2010, 91 – 107 92 Intervention. European Journal of Economics and Economic Policies former Federal Reserve Chairman Alan Greenspan and current Chairman Ben Bernanke.1 Th at policy confi guration – a focus on low infl ation plus relative neglect of asset markets – failed to prevent the build-up of massive fi nancial fragility and has been proved seriously fl awed. Now, the depth and severity of the ›Great Recession‹ provide an opportunity to reconstruct monetary policy. Th is paper challenges the conventional wisdom regarding opposition to targeting asset markets and presents a policy framework for reining in asset and credit markets. Th is framework is based on a system of asset based reserve requirements that can enhance counter-cyclical monetary policy. Th e Greenspan-Bernanke opposition to targeting asset bubbles has two components. First, there is a pure pragmatic objection that it is not possible to identify bubbles in advance. Second, there is a theoretical objection against targeting bubbles which is that explicit asset price targeting is not desirable. Part of this latter argument is that even if bubbles could be identifi ed, it would not be possible to safely pop them without exposing the economy to enormous collateral damage. For Bernanke, the problem of asset bubbles should be addressed by regulatory and supervisory measures rather than activist policy (Bernanke 2002: 2).2 Th e current paper argues against this theoretical position, and makes the case for a particular form of activist policy that has general application as part of counter-cyclical monetary policy. Th e paper begins by presenting a simple macro model that illustrates why monetary authorities should be concerned about asset bubbles, and why conventional policy may be unable to reverse their eff ects even if implemented rapidly. Not only do asset bubbles distort economic activity when they are infl ating, they leave behind damaging effects that can reduce activity long afterward. Th is provides the policy rationale for actively addressing them. Th ereafter, the paper presents a policy framework based on asset based reserve requirements (ABRR) that permits activist anti-bubble policy interventions, but does not use the tool of interest rates which impose unacceptable collateral damage on the rest of the economy. ABRR give the monetary authority additional new policy instruments that can be specifi cally targeted on asset prices, thereby avoiding the collateral damage problem and circumventing the main argument against activist anti-bubble policy. 1 Former Federal Reserve Chairman Greenspan opposed formal infl ation targeting and targeting asset bubbles (Pearlstein 2002; Greenspan 2002a and 2002b). Current Chairman Ben Bernanke favored formal infl ation targets but was against targeting asset bubbles (Bernanke 2002; Bernanke et al. 1999, Bernanke/Mishkin 1997). 2 Former Federal Reserve Governor Mishkin (2008) has made the additional argument that there is no need to target bubbles because their adverse eff ects can be nipped in the bud (i.e. cleaned up) if conventional interest rate policy is quick to respond when they burst. Th at is an empirical argument, and there are strong grounds to doubt its validity. Th e Federal Reserve was quick to lower interest rates in response to the bursting of the US house price bubble, to the extent of earning the ire of one well known economist (Buiter 2008), yet the economy has still tumbled into what has proved the worst economic crisis since the Great Depression. Palley: Asset price bubbles and counter-cyclical monetary policy 93 Lastly, the paper does not address the ›bubble identifi cation‹ argument. In this author’s opinion, bubbles can be identifi ed. Stock market bubbles can be identifi ed through measures such as cyclically adjusted stock market price/earnings ratios, while house price bubbles can be identifi ed through measures such as house price/income ratios and house price/rental ratios. Th ere are of course diffi culties and risks (Type II errors) to bubble identifi cation, but the conduct of monetary policy always involves judgment and risk. Th is even holds for rule-based policy as the rule needs to be selected and implemented. If monetary authorities can make reasonable judgments about potential output, potential growth, and expected infl ation, they can also make reasonable judgments about asset price bubbles. 2. Central bankers’ new economic model Central bankers’ opposition to targeting asset price bubbles can be understood in terms of the theoretical framework that also guides their thinking about infl ation. Th is framework has been labeled the »new consensus« macro model (Arestis/Sawyer 2006). Figure 1 provides a stylized representation of the new consensus model. Th e core logic is that the level of aggregate demand (AD) drives fl uctuations in the output gap, which in turn drive the rate of infl ation and its deviation from target (be it explicit or implicit). Th e monetary authority then responds to these deviations according to its interest rate reaction function – a form of the so-called Taylor rule – and its interest rate response causes an adjustment of AD that brings output and infl ation back in line with target. Figure 1: Th e Fed’s new model Asset prices Interest rates Exchange rates Fiscal Policy Global business conditions Business & consumer confidence Aggregate Demand Output gap proxied by actual vs. target inflation Interest rate reaction function Th e important feature of the model is that asset prices are viewed as just one of many different factors infl uencing AD. Th us, in Figure 1 asset prices enter into the funnel of AD along with business and consumer confi dence, global economic conditions, fi scal policy, exchange rates, and interest rates. According to this view, asset price bubbles are just one 94 Intervention. European Journal of Economics and Economic Policies contributing factor to AD, and are no more worthy of a central bank’s specifi c attention than is the state of business confi dence. Just as a central bank would not try to target the state of confi dence, nor should it try to target asset prices. Instead, it should manage the overall level of AD. Th is view of the economy and the resulting approach to stabilization policy can be captured by the following simple model. Output is determined by the level of AD and is given by y = E (y , iL , PA , …) and Ey > 0 , EiL < 0 , EPA > 0 , (1) where y = output, E(.) = AD function, iL = market loan rate, PA = price of assets. Equation 1 is the conventional Keynesian IS function in which AD depends positively on the level of income, negatively on the loan interest rate, and positively on asset prices. The market interest rate is determined in the fi nancial sector according to iL = iF + m , (2) where iF = the central bank’s policy interest rate (which in the US is the federal funds rate), and m = bank interest rate mark-up. Equation 2 replaces the old Keynesian LM schedule and captures the reality of interest rate determination in a world of endogenous credit money in which the central bank sets the short-term money market rate. Th e mark-up refl ects the liquidity preference of fi nancial market institutions, and can be considered a catch all for the state of fi nancial market confi dence, and attitudes toward and assessment of risk. Th e central bank chooses its target interest rate with the goal of hitting its output target, y*. Th is generates a federal funds rate of iF * = E-1(y* , m , PA , …) and diF */dy* < 0 , diF */dm < 0 , diF */dPA > 0 , (3) Th e target interest rate is a negative function of the output target (y*), a negative function of the fi nancial sector’s mark-up (m), and a positive function of asset prices (PA) and other factors positively infl uencing AD.3 Th e model is illustrated in Figure 2. A higher output target requires a lower target interest rate because the monetary authority must bring down the market interest rate to increase AD. Likewise, a higher fi nancial sector mark-up requires a lower target interest rate. Th e reason is that to obtain the market interest rate needed to hit the output target, the monetary authority must bring down the base cost of funds. 3 Th e output target can be interpreted as the full employment level of output or the level of output consistent with the monetary authority’s infl ation target. Palley: Asset price bubbles and counter-cyclical monetary policy 95 Figure 2: Th e Fed’s model IS 0 y* Interest Rate (%) Output i =i *+m LF i * F Asset prices aff ect AD by working through the common funnel described in Figure 1. Th e eff ect of an asset price bubble, as understood within the conventional paradigm, is illustrated in Figure 3. A bubble-induced increase in asset prices causes the IS to shift up. Th at induces the central bank to raise its target interest rate in order to maintain AD at a level consistent with its output target. After the bubble is over the IS shifts back down and the central bank then lowers its target interest rate. Th e underlying logic is that economic conditions are smoothly reversible. Consequently, after a bubble the central bank can engineer a return to the initial equilibrium conditions. Figure 3: Asset bubbles and the Fed’s model IS(P ) A0,.. y* Interest Rate (%) Output i * F1 i =i *+m F0 L0 i =i *+m L1 F1 i * F0 IS(P ) A1,.. 96 Intervention. European Journal of Economics and Economic Policies 3. Why the central bankers’ model is wrong Th ere are several major problems with the above bank model describing how central banks currently respond to asset bubbles. First, the model ignores the fact that bubbles generate economic distortions that have real costs. For instance, the US internet stock market bubble of the 1990s likely distorted investment by making too much capital available at too low a price to internet companies. More recently, the US house price bubble distorted economic activity by driving up house prices, thereby causing excessive residential investment. Second, the model ignores the fact that there are real costs from using interest rates to combat the infl ationary pressures unleashed by bubbles. Such costs can be termed ›blunderbuss‹ eff ects, and they refer to the adverse impacts that increased interest rates have on sectors other than those aff ected by asset bubbles. Th us, raising interest rates to counter a bubble can adversely change the composition of output, giving rise to negative long term eff ects. One problem is that higher interest rates may decrease investment spending, which in turn reduces future productivity and output. A second problem is that higher interest rates may appreciate the exchange rate, adversely impacting the trade balance and manufacturing. If the appreciation is prolonged, that can accelerate de-industrialization and increase the adjustment strains of globalization. Consequently, blunderbuss eff ects can have both shortand long-run impacts on manufacturing and growth. Another blunderbuss eff ect concerns income distribution (Th orbecke 1997). Here, higher interest rates adversely aff ect borrowers, while benefi ting creditors who receive higher interest payments. To the extent that many middle and lower income households are net borrowers, higher interest rates tend to worsen income distribution. Th at means using the interest rate tool to fi ght bubbles may compound income inequality because asset price bubbles disproportionately benefi t the wealthy, while fi ghting bubbles with interest rates disproportionately hurts net borrowers – who are generally the less wealthy. Th e third and most important omission concerns debt ›footprint‹ eff ects. Th ese footprint eff ects refer to fi nancial stock eff ects that linger after a bubble is over if the bubble has been fi nanced by borrowing. When interest rates come down after the bubble, past borrowing imposes debt burdens that can weigh down the economy. Th e monetary authority may then be unable to adequately off set the AD eff ects of these burdens because of the zero nominal interest rate fl oor.4 Th e working and impact of both debt footprint eff ects and interest rate blunderbuss eff ects can be incorporated into a modifi ed version of the above model. Now, the goods market is described by the following IS equation 4 It is worth distinguishing between debt-fi nanced asset bubbles and other asset bubbles. Th e former are associated with real estate bubbles and are particularly damaging because of the debt footprint they leave behind. Th e latter are more associated with stock market bubbles and appear to be less damaging and easier to escape. However, they also have real costs associated with distortion of investment decisions and the composition of output. Palley: Asset price bubbles and counter-cyclical monetary policy 97 y = E(y , iL , PA , B , D-1 , …) and Ey > 0 , EiL < 0 , EPA > 0 , EB > 0 , ED < 0 , (4) where B = this period borrowing, and D-1 = last period’s debt stock. Th e current fl ow of borrowing has a positive impact on AD, while last period’s debt stock has a negative impact. It is this debt stock that gives rise to debt footprint eff ects. Additionally, aggregate demand is decomposed into consumption, investment, net exports, and government spending as follows:5 E(.) = C(y , iL , PA , B , D-1 , …) + I(iL , e(iL) , D-1 , …) + G + X(e(iL)) (5) – M(y , e(iL)) , Cy > 0 , CiL < 0, CPA > 0, CB > 0 , CD < 0 , IiL < 0 , Ie < 0 , ID < 0 , Xe < 0 , My > 0 , Me > 0 , eiL > 0 where C = consumption, I = investment, G = government spending, X = exports, M = imports, e = exchange rate (foreign exchange/domestic currency), -1 = last period level. Consumption is a positive function of income, asset prices, and borrowing, and a negative function of interest payments and the level of debt. Investment spending is a negative function of the interest rate, the exchange rate, and the level of debt.6 Likewise, exports are negatively aff ected by the interest rate, which appreciates the exchange rate and lowers net exports. Imports are positively aff ected by exchange rate appreciation. Th e fi nancial sector is described as follows: iL = iF + m(D-1 , ...) and mD > 0 , (6) D = D-1 + B(dPA , …) and BdPA > 0 , (7) PA = PA-1 + dPA , (8) where dPA = change in asset prices. Equation 6 determines the loan rate as a mark-up over the central bank’s target interest rate (which in the US is the federal funds rate), but now the mark-up is a positive function of the debt stock. Th is refl ects the fact that increased indebtedness increases borrower riskiness, resulting in increased credit spreads – a feature that has been clearly visible in the current fi nancial crisis. Equation 7 determines the evolution of the debt stock, which is equal to last period’s debt plus this period’s borrowing. Th is pe5 For simplicity, the current model does not distinguish between residential and non-residential investment. Such sector distinctions can be introduced by adding separate investment functions, in which case higher asset (house) prices could spur residential investment spending. Additionally, residential investment spending would then be negatively impacted by debt footprint eff ects. 6 Th e exchange rate negatively impacts investment by increasing import competition, which reduces profi tability (see Blecker 2004). In a more complicated model the level of debt could be decomposed into household and fi rm debt. Th e former would impact consumption while the latter would impact investment spending. 98 Intervention. European Journal of Economics and Economic Policies riod’s borrowing is a positive function of the change in asset prices.7 Equation 8 determines the evolution of asset prices, with the term dPA capturing the eff ect of a bubble. Th e central bank sets its target interest rate as follows iF = iF * , (9) iF * = E-1(y* , PA , B(dPA) , D-1 , …) ≥ 0 , (10) Th us, the policy interest rate is set with an eye to hitting the output target. Th e policy rate is aff ected by asset price bubbles through their impact on borrowing and AD. Confronted by a bubble that increases AD, the central bank raises its policy rate to neutralize the bubble’s AD impact. Th e blunderbuss eff ect of interest rate policy operates via Equation 5. An asset price bubble increases AD, causing the central bank to raise interest rates. Th is has a negative impact on investment spending. It also appreciates the exchange rate, which has a negative effect on exports and a positive eff ect on imports. Such blunderbuss eff ects were clearly present in the most recent US economic expansion. Th us, as the Fed gradually raised interest rates to try to slow the house price bubble and construction boom, this contributed to a strong dollar, record trade defi cits, and weak non-residential investment spending. Th e debt footprint eff ect works through both goods markets and the fi nancial sector. Asset price bubbles increase consumption spending via the wealth eff ect and via increased borrowing. Increased borrowing raises debt, which then creates a debt footprint eff ect. Th e following period, when the bubble is over, the economy is left with a debt footprint that exerts a direct drag on spending in the goods market (Equation 5). Additionally, the increase in debt causes fi nancial institutions to increase their credit mark-up, widening the spread between the policy interest rate and the market loan rate (Equation 6). Th e net result is AD contracts directly, and the market interest rate rises, yielding a negative indirect eff ect on AD. Both types of eff ect have been visible in the wake of the bursting of the US house price bubble. From a policy perspective the danger is that the economy may get stuck in a postbubble trap, such as is illustrated in Figure 4. Th e source of the problem is the zero bound to the nominal policy interest rate. Th us, given post-bubble depressed AD conditions and higher interest rate mark-ups, the monetary authority may not be able to push its policy interest rate to a level suffi ciently low to achieve its real output target. In Figure 4, full employment requires a loan rate of iL * , which in turn requires a central bank target rate of iF * < 0 . Th at is not possible because of the zero bound, and instead the central bank must settle for a policy rate of iF * = 0 . As a result the loan rate is iL= m(.) > iL * , leaving the economy demand constrained and short of full employment. 7 If debt is decomposed into household and corporate debt this would require introducing separate loan demands for household and corporate debt, as well as introducing separate loan interest rates. Palley: Asset price bubbles and counter-cyclical monetary policy 105 8. ABRR and the euro zone ABRR have particular relevance for the euro zone and the European Central bank (ECB). Th e establishment of the euro represents an important step in the creation of an integrated European economy. Over time it should yield dividends as increased competition and lower transaction costs generate increased effi ciency. However, member countries have had to give up their own exchange rates and interest rates, and that has created problems for economic management by reducing the number of policy instruments. In particular, the ECB must wrestle with how to set interest rates when some countries are booming while others suff er high unemployment. ABRR can help fi ll this policy instrument gap. Th is is because the ABRR can be implemented on a national basis. For instance, real estate lending, which has been a major concern, is particularly suited to this. Th us, when Spain and Ireland were suff ering excessive house price infl ation, the Spanish and Irish central banks could have raised reserve requirements on mortgage loans secured by property in those countries. Th at would have raised Spanish and Irish mortgage loan rates without aff ecting rates in the rest of the eurozone. Conversely, now that Ireland and Spain are suff ering house price defl ation, they would be able to lower reserve requirements on mortgages. Nationally contingent ABRR will create incentives to shop for credit across countries. Th at means ABRR with a geographically specifi c dimension will work best when linked to geographically specifi c assets that cannot escape. Th is includes mortgage lending that is secured by collateralized property, and shares for which legal title is registered where companies are incorporated. For instance, mortgage loans are secured against specifi c real property, which determines the jurisdiction in which the loan falls and makes it diffi cult to escape compliance. More generally, jurisdictional shopping involves transaction costs. Th ose transaction costs provide a wedge that allows ABRR to create cross-country interest rate diff erentials for wide categories of assets. Lastly, jurisdictional shopping would tend to promote crosscountry fi nancial integration, which is a long-term goal of the euro project. So even here there is an upside. One possible problem with a system of ABRR is that it could raise political confl icts between the ECB and member countries. Th at suggests a two-tier system of ABRR, which would operate at both the eurozone and national levels. Eurozone ABRR policy would be controlled by the ECB, and the ECB would have the power to set ABRR across the euro zone with common requirements in all countries. National central banks would have the right to set country specifi c asset reserve requirement ratios, subject to the proviso that those requirements be no lower than the requirement ratio set by the ECB. Th is would give countries the power to set monetary policy that was tighter than that set by the ECB, but not looser. Such a system puts in place a fl oor to monetary policy that is needed to protect the integrity of the euro, but it gives individual countries the ability to pursue independent, tighter monetary policy if deemed necessary. 106 Intervention. European Journal of Economics and Economic Policies 9. Conclusion In recent years there has been debate over whether monetary policy should target asset price bubbles. Th at debate has become even more signifi cant in light of the destruction being caused by the implosion of the US house price bubble. Both former Federal Reserve Chairman Alan Greenspan and current Federal Reserve Chairman Ben Bernanke are on record as being against targeting bubbles. Th is paper has argued an opposing position. Asset price bubbles can be extremely harmful. Th at was shown by the earlier defl ation of Japan’s real estate bubble, and it is being shown again with the defl ation of the US house price bubble. Th at said, the paper argues against using interest rates to target bubbles because interest rate policy imposes unacceptable collateral damage. Instead, the paper recommends adopting a system of ABRR that can provide additional policy instruments that enable targeting asset and credit market excess without raising the general level of interest rates. Such a system would also strengthen counter-cyclical monetary policy. References Arestis, P., Sawyer, M. (2006): Interest rates and the real economy, in: Gnos, C., Rochon, L.P. (eds.), Post Keynesian Principles of Economic Policy, Cheltenham, UK: Edward Elgar, 3 – 21. Bernanke, B.S. (1983): Nonmonetary eff ects of the fi nancial collapse in the propagation of the Great Depression, in: American Economic Review, 73, 257 – 276. Bernanke, B.S. (2002): Asset price bubbles and monetary policy, Remarks before the New York Chapter of the National Association for Business Economics, New York, New York, October 15. Bernanke, B.S., Laubach, T., Mishkin, F.S., Posen, A.S. (1999): Infl ation Targeting: Lessons from the International Experience, Princeton: Princeton University Press. Bernanke, B.S., Mishkin, F.S. (1997): Infl ation targeting: A new framework for monetary policy?, in: Journal of Economic Perspectives, 11, 97 – 116. Blecker, R.A. (2004): Th e economic consequences of dollar appreciation for US manufacturing profi ts and investment: A time series analysis, Paper presented at the Post Keynesian Conference, University of Missouri, Kansas City, June 26 – 29. Buiter, W. (2008): Th e Bernanke put: Buttock-clenching monetary policymaking at the Fed, FinancialTimes.com, January 22. D’Arista, J.W., Schlesinger, T. (1993): Th e parallel banking system, EPI Briefi ng Paper, Economic Policy Institute, Washington DC. Friedman, B.M. (1999): Th e future of monetary policy: Th e central bank as an army with only a signal corps?, in: International Finance, 2, 321 – 338. Goodhart, C., Persaud, A. (2008): A proposal for how to avoid the next crash, in: Financial Times, January 31. Palley: Asset price bubbles and counter-cyclical monetary policy 107 Greenspan, A. (2002b): Issues for monetary policy, Remarks before the Economic Club of New York, New York City, December 19. Greenspan, A. (2002a): Economic volatility, Remarks at a symposium sponsored by the Federal Reserve Bank of Kansas City in Jackson Hole, Wyoming, August 30. Mishkin, F.S. (2008): How should we respond to asset price bubbles? Speech at the Wharton Financial Institutions Center and Oliver Wyman Institute’s Annual Financial Risk Roundtable, Philadelphia, Pennsylvania, May 15. Palley, T.I. (2004): Asset based reserve requirements: Reasserting domestic monetary control in an era of fi nancial innovation and instability, in: Review of Political Economy, 16, 43 – 58. Palley, T.I. (2003) Asset price bubbles and the case for asset based reserve requirements, in: Challenge, 46, 53 – 72. Palley, T.I. (2000) Stabilizing fi nance: Th e case for asset based reserve requirements, Financial markets and society, Th e Financial Markets Center, Philomont, VA, August. Pearlstein, S. (2002): Fed defends stock bubble performance: Greenspan attempts to restore his image, in: Th e Washington Post, December 22. Pollin, R., (1993): Public credit allocation through the Federal Reserve: Why it is needed; How it should be done, in: Dymski, G.A., Epstein, G., Pollin, R., (eds.), Transforming the US Financial System: Equity and Effi ciency for the 21st Century, Armonk, New York: M.E. Sharpe, 321 – 354. Th orbecke, W. (1997): Who pays for disinfl ation? Public Policy Brief 38A, Th e Jerome Levy Economics Institute of Bard College, Annandale-on-Hudson, New York, December. Th urow, L. (1972): Proposals for re-channeling funds to meet social priorities, Policies for a more competitive fi nancial system, Conference Proceedings for Federal Reserve Bank of Boston, June, 179 –189.