Cambridge Journal of Economics 2009, 33, 653–663
doi:10.1093/cje/bep036

Why don’t the bailouts work? Design of
a new financial system versus a return to
normalcy
Jan Kregel*
The innovative support measures introduced by the US Central Bank and Treasury
in response to the current crisis to bolster bank balance sheets have had little success
in restoring liquidity to financial markets. These policies mirror similar policies
employed in the 1930s in the USA and the 1990s in Japan, in both cases with little
impact. This paper identifies three policies impacting incomes rather than prices, the
assessment of system failure, and proposals for system design that were employed in
dealing with prior financial crises. That they have not been introduced in response to
the present crisis may explain why current measures have not yet had their intended
impact of restoring bank lending to the productive economy.
Key words: Liquidity crisis, Debt deflation, Bank balance sheets, New Deal financial
regulation, Zirp (zero interest rate policy), Quantitative easing
JEL classifications: E12, E32, E58, G21, G38, G19

1. Introduction
Despite the creation of a myriad of US Federal Reserve (hereafter referred to as the Fed)
special discount window facilities, unlimited swap lending by the Fed to central banks
throughout the globe and the use of Troubled Asset Relief Program (TARP) funds to
bolster bank capital there appears to be little improvement in conditions in financial
markets and it is widely lamented that the banking system is not providing lending to
support the private sector. Comparison of the response to the 1930s Great Depression and
the similar crisis in Japan in the 1990s suggests at least three factors that are absent in
current policy: (i) direct measures to support bank incomes through interest rate policy, (ii)
a clear understanding of the failures of the ‘modernised’ financial system, and thus (iii)
a clear design for the shape and structure of the financial system that is to replace it.

2. Diagnosis is more important than the cure
The prevailing diagnosis of the difficulties involved in reviving the financial system is based on
the idea of a ‘liquidity trap’. This explanation is similar to that proposed for the 1990 Japanese
Manuscript received 25 March 2009; final version received 8 May 2009.
Address for correspondence: Levy Economics Institute of Bard College, Annandale-on-Hudson, NY 125045000, USA; email: kregel@levy.org
* Levy Economics Institute of Bard College.
Ó The Author 2009. Published by Oxford University Press on behalf of the Cambridge Political Economy Society.
All rights reserved.

654

J. Kregel

crisis that followed the collapse of a bubble in equity and real estate markets eerily similar to
that recently experienced in the USA. This explanation was evoked to explain the decision by
the Bank of Japan to introduce a zero interest rate policy (zirp) in 1999, in response to the
unsuccessful attempts during the decade to force an increase in bank lending through massive
increases in bank reserves. To the frustration of the Bank of Japan, and those American
economists who supported the policy, Japanese banks simply accumulated the increased
reserves without any expansion in their domestic lending. At the same time the Bank came
under criticism from politicians who argued that their constituents’ interest income was being
decimated by the decline in the rates of interest on their postal savings accounts.
Initially, the Japanese liquidity trap referred to the creation of high powered money by the
central bank that was ‘trapped’ on the asset side of the balance sheets of the banks without
creating any expansion in deposit liabilities representing loans to business. This could also be
looked upon as a collapse of the money multiplier or of the velocity of circulation. It is thus an
interpretation within the framework of the quantity theory equation of exchange, which
would suggest that with a stable ratio of reserves to deposit an increase in reserves should
have led to a multiple expansion of loans and deposits and thus an increase in the money
supply. The aim of the policy was to increase the rate of growth of the money supply and thus
of prices, leading to expectations of inflation and an increased incentive to borrow.
When the money supply failed to respond, the central bank introduced the zirp; the
liquidity trap became a descriptive statement rather than a theoretical proposition, given
the nominal zero lower bound on interest rates. If the rate of interest is zero, by definition it
cannot be reduced.1 In this version of the liquidity trap the failure of the banks to increase
lending is due to the implicit assumption that the rate of interest that would equilibrate the
demand for loans and the supply given by the money multiplier is less than the lower bound
of zero for nominal rates.2 Indeed, if prices were falling it was possible that real interest
rates could be rising in conditions of zero nominal rates, moving the system farther away
from equilibrium. This zirp version of the liquidity trap led commentators such as Paul
Krugman and Ben Bernanke to propose that the Bank of Japan carry out a policy of reserve
expansion ‘a outrance’, what is now known as ‘quantitative easing’, in order to produce
inflation, or at least to raise inflationary expectation sufficiently to drive the expected real
interest rate negative. Indeed, Bernanke argued that such a policy, if carried on for
a sufficient period of time, would inevitably lead, as a matter of pure economic logic, to an
increase in lending and rising prices:
The general argument that the monetary authorities can increase aggregate demand and prices,
even if the nominal interest rate is zero, is as follows: Money, unlike other forms of government
debt, pays zero interest and has infinite maturity. The monetary authorities can issue as much
money as they like. Hence, if the price level was truly independent of money issuance, then the
monetary authorities could use the money they create to acquire indefinite quantities of goods
1
However, this is not the way that Keynes explained the liquidity trap. For Keynes the liquidity trap was
a price relation—the failure of the central bank to bring about a reduction in market interest rates by
increasing the supply of money. It was thus an expression of absolute or complete liquidity preference. The
public was willing to hold as much cash as the central bank would create at a constant interest rate. Keynes’s
explanation was linked to expectations of the future course of interest rates. If investors believe that interest
rates have fallen so low that they may rise by more than the square of the currently prevailing rate then the loss
in value of a coupon security would more than offset the coupon yield. In such conditions it would be rational
to sell securities at the current interest rate and hold money.
2
Although interbank deposit bid rates were negative for some periods in 1998 due to the perceived risk of
insolvency of Japanese banks and the preference for holding deposits in foreign banks operating in Japan. It
should be noted that from the time of Silvio Gesell, economists have proposed methods to make nominal
rates negative. One of the most successful was the German Bardepot.

Why don’t the bailouts work?

655

and assets. This is manifestly impossible in equilibrium. Therefore money issuance must
ultimately raise the price level, even if nominal interest rates are bounded at zero. This is an
elementary argument, but, as we will see, it is quite corrosive of claims of monetary impotence.
(Bernanke, 2000)

As an alternative method to induce inflation in Japan, it was also suggested ‘that the BOJ
[Bank of Japan] should attempt to achieve substantial currency depreciation through large
open-market sales of yen. Through its effects on import-price inflation . . . on the demand
for Japanese goods, and on expectations, a significant yen depreciation would go a long way
toward jump-starting the reflationary process in Japan’ (Bernanke, 2000, p. 160) This was
actually attempted in July 1999 but resulted only in yen appreciation, largely because the
USA was unwilling to allow the value of the dollar to rise.1
But even the zirp did not provide expansion in the money supply2 or recovery in Japan
and, as Richard Koo (2003) points out, firms may choose to pay down debt rather than
borrow even when the interest rate is negative, if this is necessary to avoid insolvency and
bankruptcy. It was only after the Japanese government introduced massive deficit
expenditure packages that the economy stabilised as this provided a more efficient way
to deleverage balance sheets. It is interesting that the Japanese government was quick to
introduce a government stimulus package in the current crisis.

3. Some measures from the Depression era
These proposals to bring recovery to Japan echo those made by Irving Fisher to resolve the
US Great Depression of the 1930s. The Fed had been created to provide a flexible
currency and its operative theory was the ‘real bills’ doctrine. As a result the issue of Fed
notes was backed by commercial loans and gold. The currency issue thus responded to the
level of activity represented by the commercial lending of banks. This worked well in an
expansion but limited the flexibility of the note issue for policy purposes in a business
decline since the only way that the Fed could increase the money supply was by discounting
member banks’ eligible assets, that is, commercial loans. But in the Depression there were
increasing commercial borrower defaults, so that money creation became procyclical. In
the absence of increased commercial lending, a change in legislation3 was required to allow
District Reserve banks to increase the outstanding supply of Reserve notes through the
purchase of Treasury securities, that is, what is now the normal policy of open market
operations. Following policies consonant with the recommendations of economists such as
Irving Fisher and members of the Chicago School4 to engineer reflation via expansion in
1
According to Richard Koo (2003), US Treasury Secretary Lawrence Summers actively opposed the
move after the Bank of Japan had spent three trillion yen without clearing the move with the USA. The
attempt to intervene in the exchange markets did, however, earn Eisuke Sakakibara, Japan’s former vice
minister for international finance, the title ‘Mr Yen’.
2
It did, however, lay the groundwork for the ‘Yen carry trade’ in which international banks borrowed yen
and invested them in higher yielding currencies in Asia and the English speaking markets, making profit from
the positive carry produced by the large interest differential. Since these operations required the sale of yen
they did not produce any increase in the domestic money supply. It also meant that whenever there were
difficulties in foreign markets or a risk of a reversal of zirp the Yen would appreciate as these positions were
closed. One such event is considered to have been crucial in the period leading up to the collapse of the Thai
baht in 1997.
3
The Glass–Steagall Act of 1932.
4
The Chicago economists favoured an increase in the money supply as the basis for a recovery in prices
but, given that this could take place only through an increase in demand for loans for productive purposes,
they argued that public spending would be more efficient than open market policy in restoring prices (see
Davis, 1968).

656

J. Kregel

the money supply, the Fed, in 1932, embarked on a policy of buying Treasury securities. The
result was a large increase in banks’ holdings of Reserve notes. However, there was little
expansion in lending, and the policy was quickly suspended, largely due to the complaints of
the banks themselves who were supposed to be the beneficiaries of the policy. The basic
reason was that since there were few qualified borrowers banks were not eager to expand
lending; since their only income was from their Treasury securities their earnings were
determined by the level of interest rates. By buying Treasuries and driving down interest rates
the Fed was rapidly reducing bank incomes to such a level that they no longer had sufficient
earnings to meet their current operating expenses. Abandoning the policy of monetary
expansion and allowing interest rates to rise was thus a means of financial stability by supporting bank incomes, even if it led Friedman and Schwartz (1971) to garland the Fed with
the failure of the economy to recover for failing to produce an expansion in the money supply.1
The expansion of the money supply by the Fed was not the only policy in support of
price ‘reflation’. The Roosevelt administration also moved to suspend the gold standard
and devalued the dollar in the belief that this would raise commodity prices and support
reflation.2 It also introduced the Agricultural Adjustment Act and the National Industrial
Recovery Act to raise prices and wages in agriculture and manufacturing industry,
providing the legislative support for firms to act as a cartel to set prices. The basic idea
behind the Roosevelt Administration’s approach, with the backing of Fisher, Viner,
Simons and others, was to use any and all means to return prices and incomes to their precrisis levels in order to allow debtors to meet their commitments. With price recovery the
system could go back to the expansion of the 1920s.3
It is telling that the Fed’s response to the crisis has been praised because it has
introduced the zirp more rapidly than the Bank of Japan in its 1990s crisis and has
eventually moved to introduce ‘quantitative easing’. However, in the absence of eligible
borrowers the only impact of lower interest rates is to reduce household incomes. As yet
there are no measures being proposed that would provide similar support to bank earnings.
The change in legislation to allow payment of interest on reserve deposits at the Fed does
not offset this impact since the interest rates are paid at a discount to the Fed funds rate.
Under zirp this means that banks receive the Fed funds interest rate of zero. The extension
of quantitative easing from expansion of the Fed’s balance sheet to the purchases of longer
term Treasury securities to flatten the yield curve was aimed at supporting household
disposable incomes. Since mortgage rates are benchmarked to 10-year Treasury rates and
there is no penalty for refinancing conforming mortgages, such a measure should increase
disposable incomes. Thus the only benefit to banks is the zero effective cost of funds, with
little possibility of increasing revenues outside of credit card balances and other
underwriting activities associated with selling the debt of undercapitalised banks. The
tightening loan standards on refinancing and the decline in interest income on household
deposits have limited any benefit from this extension of quantitative easing.
1
See Epstein and Ferguson (1984) who also note that differences in economic conditions across the twelve
Federal Reserve Districts and concern to preserve the gold backing of the currency and of the stability of
foreign deposits were also a consideration in the reversal of the expansionary monetary policy.
2
This is undoubtedly the source of subsequent recommendations made to Japan in the 1990s and
a plausible explanation of the clear decision by the USA to abandon its strong dollar policy. Market strategist
Frank Veneroso (2008) clearly outlines the similarities between Bernanke’s policy recommendations for
Japan and the conduct of Fed policy in this crisis, in particular drawing dire conclusions for the value of the
dollar.
3
Fisher’s position was supported by recognition of the impact of deflation on the real value of debt, which
could create an incentive to sell despite falling prices, and a process that Fisher called a ‘debt deflation’.

Why don’t the bailouts work?

657

At the same time the various swap arrangements (generally denominated ‘term’
facilities) in which the Fed exchanged Treasury securities for impaired assets on bank
balance sheets has had little impact on bank earnings, only on the credit quality of their
investment portfolios. More important have been the direct subsidies to banks’ costs of
borrowing through the Federal Deposit Insurance Corporation’ (FDIC) Guarantee
Program, which provides a government guarantee to banks’ unsecured securities issues.
Introduced in October of 2008, by the first quarter of 2009 some US$300 billion of
unsecured bank debt, primarily commercial paper and medium-term notes ranging from
3-months to 3-years maturity had been issued under this programme at rates equivalent to
or below prevailing rates for equivalent government securities.1
This is a fundamental difference from the policies adopted in the 1930s. It also differs
from the policy adopted by Alan Greenspan who ‘leaned against the wind’ and allowed
a persistent and sufficient spread between short rates and medium-term government
securities rates to allow banks to earn enough income from riding the yield curve to help
restore their balance sheets. The use of quantitative easing to drive down medium-term
rates prevents this solution. Indeed, those banks that have reported substantial increases in
profitability appear to have used government-subsidised costs of funds to finance their
traditional trading activities. For example, Goldman Sachs reporteds a record US$6.56
billion in revenue from trading fixed-income, currencies and commodities in the first
quarter, attributed to very high bid-offer spreads, ‘favourable competitive dynamics’, wider
margins and higher volatility (see Christine Harper, 2009).
It would thus seem clear that the basic thrust of US policy has been to restore asset
prices. This is expressed clearly in the assessment of the Fed–Treasury approach outlined
succinctly in this analysis by one of the most important fixed income managers in the USA:
The deleveraging of the shadow banking system has set ‘pawn shop’ prices on many otherwise
high-quality securities . . . The prices in the market are not indicative of the long-term value of
many of the high-quality securities in the market. . . . However, since the Treasury, the Fed and
the FDIC (Federal Deposit Insurance Corporation) do not want securities to trade at the pawn
shop bid level, they have developed programs intended to support prices . . . effectively stepping
around the banking system to create a sort of government shadow bank to provide financing for
asset-backed securities (ABS) . . . with the aim of supporting prices on ABS and incrementally
nudging the prices up closer to their intrinsic value. This has created an opportunity to buy at
attractive prices and potentially earn an attractive yield while also being in the position to
potentially pick up capital gains if spreads tighten as the full force of Uncle Sam’s balance sheet
comes into play. (Paul Reisz, 2009)

The support to bank profits is to come not from income on net interest margins but from
trading in an environment in which prices are irrationally depressed, from providing
funding to restore the balance sheets of financial institutions who can then purchase the
assets from deleveraging institutions at prices below ‘intrinsic’ value. Whether this policy
will work depends on the belief that the excessive liquidity premium has driven values
below their intrinsic levels. But, even if this is correct, it does not provide a long-term
business model to ensure bank profitability.

4. The change in financial structure......
Although the New Deal policies included direct income support through unemployment
insurance and direct employment measures such as the Public Works Administration and
1

See FDIC data at http://www2.fdic.gov/qbp/2008dec/tlgp6c.html and Tyler Durden (2009).

658

J. Kregel

Civilian Conservation Corps, active Keynesian-style deficit spending in support of
incomes was only introduced after the ill-fated decision in 1937 to balance the budget
was combined with increasing interest rates to produce another downturn. On the other
hand, the approach included another crucial element absent from present discussions: A
fundamental change in the financial structure through a series of regulatory measures and
the creation of a series of new regulatory institutions. Since these changes are relatively well
known they need not be rehearsed here, except to note the absence of such measures in the
current situation, or even any discussion of the desired post-crisis structure of the financial
system. Indeed, policy has changed course so frequently that complete uncertainty prevails
over what kind of financial system will emerge. Despite the affirmation that the government
does not want to run financial institutions and that nationalisation is not an option, the
introduction of retroactively binding legislation to fix compensation in the financial sector
can only confirm the uncertainty and the absence of any clear strategy concerning
reregulation and reform of the system.

5. .....A key missing element in current policy
This point is of more general importance to the issue of why banks have not increased
lending despite the massive bailout package. The US banking system in the aftermath of
the 1999 Financial Modernization Act was a major departure from the principles that
produced the New Deal banking legislation. The new system gave financial institutions the
role of intermediaries between borrowers and capital markets: what Minsky called money
manager capitalism. The banks reduced the loans held at risk on their own balance sheets
to a minimum in order to conserve capital and thereby increase their rates of return by
maximising the fees and commissions earned from this management and intermediation
role. It should not be surprising that banks have not increased lending in response to the
recovery measures—in the new system they had already ceased to be the most important
lenders in the system. Losses from real estate lending had already reduced their capital,
making it difficult to resume their own lending, and the fact that capital market institutions
had stopped buying the loans the banks had originated because of a lack of transparency
concerning their risks meant that credit to the entire system ceased.
The second element that supported bank earnings was leverage. The institutions that
provided that leverage have disappeared with the crisis, along with the markets for the
instruments that were used to create it. In addition, the exhortations from regulators to
reduce leverage have further reduced the availability of credit to the private sector. But it is
not so much the failure of banks to lend, which they did not do in any case, as the
breakdown of the financial structure created by the 1999 Act that has fed the crisis. A
return to normal would mean recreating that system, and this would mean laying the
groundwork for the next crisis. Fundamental structural reform has to be part of any
successful policy to restore financing to the productive sectors of the economy.

6. Why the current system failed
To understand this point it is important to recognise how the now collapsed system of
‘originate and distribute’ differed from the traditional ‘originate and fund’ system. Under
the New Deal financial structure bank loan officers would originate loans in the form of
new deposits and then the reserve desk would have to find the deposits or interbank lending
to reserve them at the statutory reserve ratio. If the system came up short, the Fed would

Why don’t the bailouts work?

659

have to provide the reserves. However, for any individual bank secondary reserve
assets—liquidity cushions—would be held that could be converted to reserves if there
was a shortfall, either because of an excessively exuberant loan officer or because of
a decline in the quality of loans and an increase in charge offs. The loans that were initially
funded by a creation of a bank deposit liability thus represented an unfunded liability that
had to be hedged by the bank through its liquidity policy. Or, as Hy Minsky put it:
Banking is not money lending; to lend, a money lender must have money. The fundamental
banking activity is accepting, that is, guaranteeing that some party is creditworthy . . . A bank
loan is equivalent to a bank buying a note that it has accepted . . . When a banker vouches for
creditworthiness or authorizes the drawing of checks, he need not have uncommitted funds on
hand. He would be a poor banker if he had idle funds on hand for any substantial time . . . Banks
make financing commitments because they can operate in financial markets to acquire funds as
needed; to so operate they hold assets that are negotiable in markets and hold credit lines at other
banks. The normal functioning of our enterprise system depends upon a large array of
commitments to finance, which do not show up as actual funds lent or borrowed, and money
markets that provide connections among financial institutions that allow these commitments to
be undertaken in good faith and to be honored whenever the need arises. (Minsky, 2008, p. 256).

In the world of securitised assets that was created after 1999 there was little or no holding
of negotiable assets against a loan commitment, no visible backup credit lines, no need for
money market funding. The loans were sold or were packaged together with other loans in
trusts and then sold (often presold) to another arms-length securities institution classified
as a variable interest entity1 organised to issue its own capital market liabilities that were
used to purchase the bundles of bank-originated assets. These special purpose vehicles, or
special investment vehicles, were created to transfer the risk of first loss from the bank to
the nominal owners of the trust—the capital market investors—thus removing the loans
from recourse against the issuing banks and eliminating the need to hold capital against
them as well as reducing the need for secondary sources of liquidity.
But not only was the capital backing removed in this structure, the function of the
reserve desk officer was replaced by the financial engineering that produced the liability
structure of the shares in the special entities that were sold on a non-recourse basis to
capital market buyers. As has now been widely noted, this eliminated the normal due
diligence process of the loan officer in judging the credit worthiness and quality of the
borrower and shifted the due diligence process to assessing the probability of default of the
special purpose entity. No one was tasked with the assessment of the quality of the assets
that were held by the entity. And even the due diligence on the structure was outsourced to
private rating agencies that clearly had neither the same interest as a loan officer nor the
final buyer, but rather the interests of the issuing bank that was paying the fees (Kregel,
2008).
But, more importantly the liquidity cushion of secondary reserves and access to market
financing that was normally held by a bank under originate and fund disappeared in the new
system. There was a cushion, provided through overcollateralisation or credit enhancements
1
Under US Financial Accounting Standards Board regulations a variable interest entity (VIE) is an armslength institution in which the investor holds a controlling interest not based on majority control of voting
rights. It is closely related to the concept of a special purpose entity. If the investor is the primary beneficiary of
the VIE it must report it on its consolidated balance sheet. A VIE meets one of three criteria: (i) the equity-atrisk is not sufficient to support the entity’s activities (e.g., the entity is thinly capitalised, the group of equity
holders possess no substantive voting rights, etc.); (ii) the equity-at-risk holders cannot control the entity; (iii)
the economics do not coincide with the voting interests (commonly known as the ‘anti-abuse rule’). The
classification of VIE determines whether or not it is consolidated on the balance sheet and thus whether
capital must be held against it.

660

J. Kregel

from bond guarantee insurance companies or credit default swaps, but these cushions were
not there to provide liquidity to the structure, they were there simply to provide the
justification for the investment grade credit ratings on the senior liabilities. Thus, the
movement of the loans off the banks’ balance sheets not only reduced the capital backing of
outstanding loans, it also eliminated the liquidity cushions behind the loans.
Finally, a portion of the subprime and Alt-A loans that were written with option
adjustable rates in which the borrower could choose the interest payment, or adjustable
rates with a predetermined reset from a below-market rate to market rates, were
programmed to become insolvent at the reset date in the absence of the ability of the
borrower to sell the house at a profit since the borrowers income could not cover the higher
interest payments after reset. To simplify, the secondary liquidity that was once provided by
a bank’s portfolio of Treasury securities, the safest assets traded in the most liquid market,
was replaced by the liquidity of the second-hand real estate market, one of the least liquid
markets in the financial system.
But there was an additional perverse impact. The bond insurers and issuers of credit
default swaps that provided the credit enhancement to these structures were even less well
capitalised and had even lower liquidity cushions than the banks or the structures that they
were insuring. Thus, as real estate prices started to decline and it became clear that the
non-rated equity and lower-rated intermediate liability tranches of the structure would be
impaired, questions arose about the creditworthiness of the AA and AAA senior tranches
and rating agencies started a process of downgrading their credit ratings on these tranches
of the structures. This meant that the sellers of protection through credit default swaps on
the structures had to provide additional margin to the buyers of credit protection, and the
monoline insurers who had provided guarantees of payment in the case of default also had
to increase their margin payments. In addition this had a knock-on effect on a whole series
of other assets whose ratings were determined by that of their major monoline insurer.
Thus the very institutions they were meant to provide liquidity insurance to the structured
vehicles themselves created an additional demand for liquidity in a system that had more or
less eliminated all of its traditional liquidity cushions. In normal circumstances it is the
liquidity crisis that produces the need to sell position to make position generating the
decline in asset values that leads to insolvency. In the sub-prime crisis it was the recognition
of insolvency of the securitised structures that set off a rush for margin liquidity that
engulfed the entire system.
At the same time, the rising loan to value ratios and the failure to verify income of
borrowers meant that the liquidity cushion that is normally provided by the owner’s equity
in the home and own wealth or income coverage was also negligible. When house prices
stopped rising it implied insolvency for the borrowers and also for the special purpose
entities holding the mortgage collateral and with it the liquidity provided by overcollateralisation of the structures. But these structures never possessed a traditional
liquidity cushion in the sense of liquid assets to be sold in the market. It is not surprising
that there were difficulties in finding reliable market prices for the collateral and thus for
the liabilities of the mortgage securitisations. Equally important, since the households had
no liquidity cushion other than the real estate market, it meant insolvency for the
borrowers at the reset date for the mortgages, if not before.
Had this readjustment of prices been restricted to the buyers of the securitised mortgage
entities’ liabilities and the underlying sub-prime borrowers, the collapse would have
produced a loss of wealth for the former and a loss of wealth for the latter, with the
possibility of the impact of the wealth effect on consumption and the level of activity.

Why don’t the bailouts work?

661

A short, sharp slump. But, this was not the case. Banks had warehoused, or held as
investments, the senior investment grade liabilities on their investment books. They had
also provided liquidity guarantees to the entities, as had other financial institutions, many
of whom held them on a highly leveraged basis. Once the securitised entities became
insolvent, they produced a direct negative impact on banks’ balance sheets, either because
they had to provide additional liquidity or because the losses had to be taken against bank
capital in conditions in which there were no secondary reserve cushions available. Joined
with the demand for liquidity to provide margin on credit enhancements this combined to
produce what Minsky, following Irving Fisher, called a debt deflation as it became
necessary to sell position to make position. In a market where there are no buyers, there is
by definition no liquidity, and a fortiori, no market price. It became impossible to value the
assets of any financial institution and thus impossible to ascertain its creditworthiness as
a counterparty. Thus, not only did lending to the private sector come to a halt, but lending
amongst financial institutions, which normally supports liquidity, also came to a halt. This
was simply exacerbated by the actions of the Fed in the Bear Stearns, Lehman and AIG
responses in which there was no clear principle to determine who would be supported and
who would be allowed to fail. Since even a secured loan to an institution that eventually
declares bankruptcy means loss of the collateral due to clawback provisions; if every
institution is a potential Lehman Brothers no one bank will lend to any other bank, leaving
the entire provision of liquidity to the only secure lender: the Fed.
Thus while the bailouts of financial institutions have acted to prevent insolvencies from
producing bankruptcy, they have done little to increase either the willingness, or the ability
of banks to lend. And this is independent of the decline in qualified borrowers that has been
produced by the sharp decline in the overall level of activity. The crisis has destroyed the
‘modernised’ lending mechanism based on leveraged securitisation for the private sector
financial and business sector without putting any clear mechanism in its place. Thus, more
important than a mechanism to remove the impaired assets from bank balance sheets, what
is required is a clear roadmap of what the design of the new financial system should be.
Under the New Deal, in a very short space of time between 1933 and 1935, this was
provided. Nearly the same amount of time has passed and still there is no clear vision of
what the new New Deal will be. It is also clear that the policy of attempting to return prices
to pre-crisis levels, to save the existing system, has not worked. But, this seems to be the
prevailing approach to the resolution of the crisis: if only the impaired assets can be held
long enough they will recover their value and commitments can be met.

7. The final lesson from the New Deal
Thus, comparing the current response to the crisis it appears not to have absorbed the
importance of the negative impact of low interest rates on incomes, while it has accepted
the Fisherian idea, resurrected in the Japanese crisis by Krugman and Bernanke, of
resolution through zirp and quantitative easing to restore asset and goods prices. These are
policies that did not work in the USA in the 1930s or in Japan in the 1990s. However, this
approach to ‘restoring normalcy’ explains the absence of the most important aspect of the
New Deal—the rapid assessment of the causes of the failure of the existing financial system
and the design and introduction of a new financial structure that eliminates the causes.
Instead the restoration of asset values and the removal of ‘impaired’ assets from the balance
sheets of institutions to restore them to health suggest that they will be able to operate
much as they have in the past. This seems impossible.

662

J. Kregel

This leaves the final aspect of the Japanese and New Deal resolutions of crisis. Despite
the common opinion to the contrary, Roosevelt ran on a platform that accused Herbert
Hoover of being a profligate spendthrift and promised to balance the budget in both his
first and second election campaigns. It is somewhat ironic that Ronald Reagan ran his
campaigns on quotations from Roosevelt’s speeches in support of balanced budgets. As
noted above, it was Fisher rather than Keynes that dictated New Deal policy. Keynesian
style deficit spending was only adopted as an emergency policy after the increases in
taxation introduced in 1935–36, which look disturbingly similar to those discussed by the
current administration, produced a downturn in 1937. The discussion of the current
stimulus bill has centred on the ability of spending to create employment. But, as
important as increasing employment may be, if the above analysis is correct and the basic
problem is a financial system that attempted to function with reduced liquidity cushions,
and has virtually eliminated liquidity from the system, and it is this lack of liquidity that is
producing the contraction of the productive sector due to lack of financing, the initial focus
of any government expenditure should be directed to providing income to cover losses
sustained by banks and households and restoring system liquidity. Increasing incomes
should be directed not only to increasing employment but also to covering losses and
restoring liquidity. If the level of government expenditure is sufficiently high and
sufficiently stable to provide full employment, liquidity will be automatically provided to
the system via stable income and stable sales receipts, which assure that debts can be
liquidated through sale of assets. The best way to reduce liquidity demands is by assuring
that they are provided by the steady full employment cash flows of firms and incomes of
households.
Thus, the basic difference is whether liquidity and prices can be restored through an
increase in high powered money at zero interest rates or through increased government
spending that increases incomes, expenditures and profits. Whether it comes from creation
of employment or not is, in the first instance, unimportant. But, it is important that there is
increased employment creation and the associated increase in income, whether for banks
or for households. This was the point of Keynes’s recommendation that it would be
sufficient to bury jars full of bank notes and allow people to dig them up. Only when banks
have sufficient incomes will they be able to restore their capital and recommence lending.
Only when households have sufficient income to pay down their debts will they be able to
recommence spending. Up until now it has been the Fed that has been willing to operate on
banks’ balance sheet, but only by swapping one asset for another, not by increasing bank
earnings. As noted there are subterfuges such as those operated by Greenspan in the 1990s
to do this. The best way is to increase incomes sufficiently so that households can meet the
debt service on their loans out of income and firms have sufficient income to meet their
borrowing. As Keynes himself noted, one of the simplest ways to offset an increase in
liquidity preference, that is, the increased demand to hold cash rather than lend it to
finance productive enterprise, is to increase the earnings of households and businesses.
Building hospitals and bridges would be nice, but if it is impossible to convince politicians
that this is a good thing, then we have to support employment of bank note jar
archeologists. As Keynes noted in this regard, ‘It is not quite correct that I attach primary
importance to the rate of interest . . . I should regard state intervention to encourage
investment as probably a more important factor than low rates of interest taken in isolation’
(Keynes, 1980).
However, as Hy Minsky has pointed out, a better way to solve both the liquidity and the
income problem would be by channeling government expenditure through an employer of

Why don’t the bailouts work?

663

last resort programme. This would provide a direct increase in incomes without creating
the additional financial layering that would be associated with supporting additional
investment expenditure. Indeed, one of the major causes of the current difficulties is that
household consumption has not been financed by increasing real wages in step with
productivity, but rather through increased household borrowing that fuelled the increase in
incomes to the financial sector. In simple terms, the current crisis could have been avoided
if increased household consumption had been financed through real wage increases
following productivity and if financial institutions had used their earnings to augment bank
capital rather than increasing salaries and bonuses for their proprietary traders. This
suggests that what is needed is reform that provides financing for productive investment
that increases labour productivity and that is reflected in increased real wages of
households with the increased earnings of the financial system going to increasing bank
capital.

Bibliography
Bernanke, B. 2000. Japanese monetary policy: a case of self-induced paralysis? in Posen, A. S.
and Mikitani, R. (eds) Japan’s Financial Crisis and Its Parallels to U.S. Experience, Special
Report 13, Institute for International Economics, Washington, DC
Davis, J. R. 1968. Chicago economists, deficit budgets, and the early 1930s, The American
Economic Review, vol. LVIII, no.3, Part I, 476–82
Durden, T. 2009. Bailoutspotting (Or The Search For The Great Financial Methadone Clinic),
14 March, available at http://zerohedge.blogspot.com/2009/03/bailoutspotting-or-search-forgreat.html [date last accessed 3 June 2009]
Epstein, G. and Ferguson, T. 1984. Monetary policy, loan liquidation, and industrial conflict:
the Federal Reserve and the open market operations of 1932, The Journal of Economic History,
vol. 44, no. 4, 957–83
Friedman, M. and Schwartz, A. J. 1971. Monetary History of the United States, 1867–1960,
Princeton University Press, pp. 888
Harper, C. 2009. Goldman Sachs’s $100 Million Trading Days Hit Record (Update1),
Bloomberg online, 6 May, available at http://www.bloomberg.com/apps/news?pid5
20601087&sid5a7HGVAn8w73Y&refer5home [date last accessed 3 June 2009]
Keynes, J. M. 1980. Letter to J. Wedgwood, 7 July 1943, reprinted, pp. 350–1 in Moggridge,
D. E. (ed.) The Collected Writings of John Maynard Keynes, Vol. 27, London, Macmillan
Koo, R. 2003. Balance Sheet Recession: Japan’s Struggle with Uncharted Economics and Its global
Implications, Singapore, John Wiley
Kregel, J. 2008. Using Minsky’s cushions of safety to analyze the crisis in the U.S. subprime
mortgage market, International Journal of Political Economy, vol. 37, no. 1, 3–23
Minsky, H. 2008. Stabilizing an Unstable Economy, McGraw-Hill, New York
Reisz, P. 2009. Opportunities to increase yield by moving from cash to ‘near cash’, Pimco
Viewpoints, April, available at http://www.pimco.com/LeftNav/Viewpoints/2009/ Viewpoints1
April120091Reisz1Opportunities1to1Increase1Yield1by1Moving1from1Cash1to1
Near1Cash.htm [date last accessed 3 June 2009]
Veneroso, F. 2008. Let Us Hear IT from Helicopter Ben, Veneroso Associates, Portmouth, NH.
www.venerosoassociates.com