Memorandum from Jan Toporowski
ECONOMICS DEPARTMENT, THE SCHOOL OF ORIENTAL
AND AFRICAN STUDIES, UNIVERSITY OF LONDON AND THE RESEARCH CENTRE
FOR THE HISTORY AND METHODOLOGY OF ECONOMICS, UNIVERSITY OF AMSTERDAM
It is utopian to consider the issue of international
banking regulation in isolation from global macroeconomic imbalances
and their financing. It highlights the role of asset inflation
in giving temporary stability to financial systems, at the cost
of severe damage to banking systems when that inflation ceased
[paragraphs 1-3]. Prudent banking policy needs to be supplemented
by accommodating fiscal policy, if only because (domestic) government
debt is a stabilising factor in bank balance sheets [4-6]. But
government re-capitalisation of banks may be a wasteful use of
government resources . The evidence takes issue with the view
that the markets or independent regulators can regulate financial
markets effectively or well [7-8]. The evidence criticises more
extensively the view that additional equity capital cushions banks
against risks in their lending. In fact such additional capital
reduces the equity capital available to non-financial firms. Such
firms then have to rely on debt financing. Such "forced"
debt financing discourages productive investment, weakens the
liquidity of the non-financial business sector, and in this way
degrades the quality of bank assets [11-18]. Over-capitalisation
of banks is exacerbated by cross-border lending, which under present
arrangements requires more capital to be put aside . The evidence
concludes by suggesting an alternative system in which cross-border
lending to the private sector abroad is supported by additional,
contingent lending commitments to foreign governments that may
be required to refinance the foreign debts of their private sector
I. Introduction: Banking Depends on Macroeconomic
1. Institutions, how they function together
in markets, and how macroeconomic imbalances are accommodated
are more important than principles of banking regulation abstracted
from the actual intermediation functions of banks. The present
crisis is not due to deregulation (which had occurred in 1980s)
but precisely because commercial credit was allowed to take up
the slack in the economy that arose after the dot.com bubble burst
in 2000. Commercial credit was able to support consumption through
housing market inflation, and corporate liquidity through capital
market inflation, at the cost of rising indebtedness that emerged
as excess debt when housing and financial asset prices fell.
2. A second disadvantage of discussing bank
regulation abstracted from actual conditions of financial intermediation
is the unreality of any discussion that starts with general principles
of regulation when banking and financial markets are gripped,
or at least driven, by crises in the U.S. and U.K. that are notable
for their durability rather than their tractability. Much of the
recent discussion among academic economists of "optimum"
regulation in a private sector banking system, hardly addresses
the current prospects for banking. The current difficulties are
largely regarded as policy problems, rather than regulatory ones.
Especially in the most financially advanced countries, banking
will be in the state sector for a long time to come and will be
affected for decades, if not years by the measures taken to deal
with the crisis. Discussions of bank regulation that abstract
from the banking structures being currently worked upon are either
frankly utopian or else implicitly nostalgic for a recent, but
receding, past when macroeconomic imbalances were accommodated
by asset inflation, rather than the efficiency of deregulated
3. The durability of the present crisis
suggests that the discussion should be centred upon not so much
financial regulation, but financial reconstruction. Such financial
reconstruction has to take place within a broader programme of
the reconstruction of our economic institutions, whose aim must
be to reduce the dependence of global economic activity on financial
market conjunctures. Such dependence has been embedded in the
financialised economies, such as the U.S., the U.K., and Iceland,
through funded pension schemes, the use of privatisation as a
source of government revenue, and the increasing use of the housing
market to support household consumption, while the non-financialised
economies, such as Germany or China, through their increasing
direct or indirect reliance on exporting to financialised economies,
thereby acquired an indirect dependence on financial inflation.
Economic activity must be reorganised to make the real economy
more stable and financial markets more boring, so that enterprise
is focussed on innovations that enhance welfare, rather than financial
4. The macroeconomic preconditions for financial
stability must include appropriate fiscal accommodation. It is
important to remember that the stability of banking and financial
markets after the Second World War was not only due to regulation,
but also due to the overhang of government debt largely held by
banks and kept liquid by central banks. Such government debt operations
must be part of any effective stabilisation of banking markets.
To some extent this is already happening as government debt expands
rapidly to refinance banks; stabilising banks by distributing
among financial balance sheets government paper backed, in effect,
by claims on banks.
5. An important limitation of this emergence
of the government as financial intermediary to bankers is that
government credit operations are being used to support credit
("finance is financing finance") rather than activity
in the real economy, as happened in previous government debt expansions.
There is a benefit in that it facilitates "quantitative easing",
that is the purchase from banks of government paper by central
banks to provide commercial banks with additional reserves. But
the benefits of quantitative easing have always been greatly exaggerated:
It did not have any significant economic effect in Japan when
it was tried at the end of the 1990s, when the ratio of Japanese
Government debt to Gross Domestic Product was approximately 150%
(the U.S. and U.K. government debt to GDP ratios currently stand
at 60% and 50% respectively).
6. A particular difficulty, which may have
added to the fragility of banking systems in the U.S. and the
U.K. and which today limits central bank operations with commercial
banks, is the concentration of government paper in the reserves
of central banks and long-term investment institutions (insurance
companies and pension funds). The clear preference among long
term investment institutions, partly for regulatory reasons, for
government paper means that the return of ailing government-supported
banks to the private sector will be limited by weak demand in
the capital market for shares or common stocks. This means that
banks are likely to stay in the public sector for a while. (This
is further discussed in the next section of this evidence.) The
discussion on bank regulation will remain utopian as long as it
ignores the distinctive functions that public sector banks must
7. On the agenda of discussions about financial
reconstruction has to be elimination of so-called "independent"
central banks and regulators. One of things wrong with our institutions
is the narrowing of the policy agenda to consumer price inflation
and employment. This allowed regulators to tolerate inflation
in the financial markets and get away with it, while delegating
awkward questions of regulation to an abstracted market in which
"rational" agents prevail, or independent regulators
who had no overview of how banks were functioning in the economy
as a whole. The crisis has brought home to everyone what was apparent
in the 1930s, namely that in a capitalist market economy in general,
and in financial markets in particular, the "rational"
generation of cash flow does not result in any kind of social
equilibrium or optimum.
8. More importantly, in the moment of crisis,
it is most obvious that the previous cash flows generated by inflation
in asset markets concealed structural imbalances in those markets.
Delegating regulation to market participants, or even to those
experienced in generating cash flows from asset markets, ie, so-called
"expert practitioners" means delegating to those acting
under the influence of illusions created by those cash flows.
Among those illusions is the belief, now widely held among banking
experts that the banking system's exposure to macroeconomic hazard
arises merely out of inter-bank financial transactions which leave
banks with sizable exposures to default by a single bank (see,
for example, the so-called "Geneva Report", Brunnermeier,
M., Crockett, A., Goodhart, C.A.E., Persaud, A.D., and Shin, H.
(2009) The Fundamental Principles of Financial Regulation,
Preliminary Conference draft Geneva: International Center
for Monetary and Banking Studies). Such a definition of "macro-prudential
regulation", and the concern for securing the survival of
"systemically important" financial institutions (ie,
institutions whose default would have an adverse impact on the
balance sheets of a large number of banks) reflects the "capture"
of financial regulators by the institutions which they are supposed
to regulate. Financial inflation expands cross-holdings of each
others assets by banks in liquid markets whose cash flow creates
an illusion of successful credit operations. Independent regulators
will always incline to deregulation at such times. The answer
is to have publicly accountable regulators imbued with an ethos
of public service capable of auditing market structures and market
processes, rather than practitioners whose expertise in individual
balance sheet restructuring blinds them to market structures and
imbalances. It is vital that a public, democratically-accountable
authority be able to modify banking and financial market regulation
in line with changing macroeconomic conditions, and the emergence
of structural imbalances in the economy or in the financial markets.
The need to coordinate such regulation with the availability of
government paper to stabilise bank balance sheets suggests that
this regulation should be done by an agency associated with H.M.
II. Capital Adequacy Requirements
9. Conventional wisdom since the 1980s has
it that banks can be made more secure by having capital that is
sufficient to meet a decline in the quality of banks' assets (loans
or bonds). This thinking was reflected in the Basle Agreement
of 1988, laying down minimum capital requirements in relation
to the supposed riskiness of assets. Apart from the fact that
there is considerable uncertainty about the precise riskiness
of assets, the strategy of securing bank stability by capital
adequacy is based on a fallacy of composition: what is good for
one bank is not necessarily good for all banks taken together.
10. Essentially, the strategy is based on
a microeconomic presumption that each bank can determine its liabilities
(ie, the scale and distribution of its liabilities between deposits
and capital) without affecting the liabilities of other banks
and firms in the economy. This is a fallacy because the process
of issuing capital or liabilities is subject to two constraints.
The first of these is the balance sheet constraint that each bank's
or firm's liabilities must be some other firm's or bank's assets.
Secondly, even if one assumes that the price system (ie, the return
offered and accepted on liabilities and assets) will allow all
banks and firms to issue the kind of liabilities that they wish
to have in their individual balance sheets, there is no guarantee
that their assets will generate sufficient income to allow each
individual bank or firm to make those payments on their liabilities.
11. In practice banks cannot determine their
capital without affecting the availability of capital for other
economic enterprises. If we exclude the possibility of bank holding
companies and bank capital cross-holdings (see below, paragraph
20) the supply of capital today in the financially advanced markets
of the OECD countries is principally determined by the cash flow
and the respective liability structures of other financial intermediaries,
such as pension funds and insurance companies. Given a certain
capacity on the part of other, non-bank, financial intermediaries
for purchasing equity, a regulatory requirement to increase bank
capital reduces the amount of capital available for non-financial
firms. If non-financial firms are unable to raise the amount of
equity capital that they need, they are obliged to raise capital
through the issue of debt instruments in the form of corporate
bonds or company paper. In this way, stabilising banks by raising
capital requirements becomes a way of "forcing" companies
12. The "enforced indebtedness"
of companies is a crucial factor in bringing about financial crisis
and recession. Both the 1929 Crash, and the Crash of 2007-08 were
preceded by rising capital issues by financial intermediaries
and the growing indebtedness of non-financial firms. The response
of non-financial firms to rising levels of debt is to reduce their
(productive) investment in fixed capital. Such falls in fixed
capital formation are a key factor in bringing about recession
in the real (non-financial) economy and, in the case of the 1930s
(or in Japan after 1992) prolonging that recession into economic
stagnation and depression.
13. The "crowding out" of the
non-financial business sector's demand for capital also raises
the cost to them of the eventual capital that they raise. Because
banks have to satisfy a regulatory requirement for capital they
are willing to pay whatever price it takes to get the capital
onto their books. This "inelastic" demand for capital
means that banks may end up promising a higher return on the capital
that they raise. Non-financial firms wishing to raise equity capital
must then match or offer even higher returns on the capital that
they wish to issue. This is reflected in the rising yields on
corporate equity issues since 2000.
14. The other way in which banks can accommodate
demands for higher capital ratios (ie, ratios of capital to assets)
is by securitisation, ie, by packaging up bank loans as bonds,
and selling the resulting bonds to other financial intermediaries
(insurance companies and pension funds). This raises banks' capital/asset
ratios by reducing the amount of risky assets on those banks'
balance sheets. Securitisation has recently had a bad press because
of its role in the "business model" of failed banks,
such as Northern Rock. But there is also another channel by which
securitisation has contributed to financial fragility. This is
through the sale of loan-backed bonds to other financial intermediaries.
Such sales effectively reduce the pool of capital that non-financial
firms can raise in the markets. This is another means of "enforced
indebtedness" of companies.
15. "Enforced indebtedness" increases
the financial fragility of the economy as a whole. Firms which
would have preferred to finance themselves with equity capital
find themselves holding levels of debt which over their planning
horizon they would prefer to reduce. They can accommodate this
excess debt in one of two ways. Firms can hold larger amounts
of liquid assets (bank deposits, foreign currency deposits, short
term bills). But this means that capital which they have issued
is `wasted' by being held as financial assets, rather than being
applied productively to the expansion of output or fixed capital.
Alternatively, firms can reduce their fixed capital investment
in order to build up those liquid assets. Either way, productive
investment ends up being less than it would otherwise be. If sufficient
firms succumb to indebtedness in this way, the economy moves to
a lower growth trajectory. Fixed capital formation is a key factor
in the liquidity of firms, with higher investment in fixed capital
being associated with higher retained profits of firms. Lower
investment in fixed capital therefore reduces the liquidity of
firms and their ability to service their debts. Indirectly, therefore,
the company indebtedness enforced by raising capital requirements
for banks itself may cause a decline in the quality of bank assets.
16. Looking at this issue over the course
of the business cycle, it is easy to see how regulatory bank capital
requirements may make fluctuations in economic activity even more
extreme. Given even moderate business fluctuations, it is prudent
for non-financial firms to raise additional equity capital as
a boom proceeds because the longer the boom lasts, the closer
is the eventual recession, and therefore the more likely is a
fall in the return on the productive assets of non-financial firms.
If, however, banks are increasing their issue of equity capital,
then non-financial firms may find themselves unable to issue additional
equity or relying on debt finance. In this way, when the recession
comes, it is made worse by the greater indebtedness of companies.
In the recession the returns on all assets, including those in
bank portfolios, deteriorate. To ensure their survival non-financial
firms should now be converting debt into equity. But since the
deterioration of their assets is obliging banks to raise more
equity capital themselves, this bank capitalisation reduces the
already diminishing pool of equity capital available to non-financial
17. More recently there have been proposals
to stabilise bank balance sheets by requiring banks to raise the
capital to risk-weighted assets ratios over the course of an economic
boom. The reason for this is that the present fixed capital to
risk-weighted asset ratios is pro-cyclical in the sense that,
in the course of an economic boom, the liquidity and net worth
of balance sheets increases. A fixed capital ratio would therefore
fail to discourage banks' risky lending and may even encourage
it because the risks would be less apparent in the boom. The policy
of requiring banks to raise capital ratios as a boom proceeds
is sometimes referred to as "dynamic provisioning".
18. There are three objections to this dynamic
provisioning approach. In the first place the risks that are supposed
to evoke higher capital ratios are rather nebulous, and evidenced
only by defaults in a recession if it comes, so that it becomes
difficult to enforce higher capital ratios if the boom is prolonged.
Secondly, the approach would require banks to drain the available
pool of equity capital even more rapidly than under fixed capital
ratios, causing a corresponding greater indebtedness on the part
of companies, and forcing up even more the cost of capital that
non-financial firms could issue. Far from stabilising banks and
the economy, dynamic provisioning would destabilise the economy
and the credit system, including banks, by indebting companies
more rapidly in a boom, or else discouraging company investment,
and by these means increasing the risk of companies' default on
their debt. The third problem arises in an international setting
in which some banks may have a diversified portfolio of assets
spread across countries, some of which are in a boom, and some
of which are in recession. Such banks would on balance have a
much more stable portfolio of loans and could reasonable argue
that they should not be required to increase their capital ratios
as much as banks exposed only to countries in recession.
19. A far more effective form of dynamic
provisioning would be to require companies (and households with
mortgages, if we are to extend our considerations to household
debt) to increase the equity capital that they have as a boom
proceeds. This would make companies' remaining debt more manageable
in the face of a recession, and thereby reduce the probability
of default on that debt. The latter in turn would improve the
quality of bank assets.
20. So far in this evidence consideration
has only been given to the issue of bank capitalisation through
the sale of equity capital to financial institutions such as insurance
companies and pension funds. The sale of equity capital to private
individuals is not an effective means of raising capital because
few private individuals are wealthy enough to be able to hold
significant quantities of capital on the scale required by banks.
However, there is another means of raising bank capital, namely
that of bank holding companies, or cross-holdings of bank capital.
A bank could issue equity capital which might be held by a holding
company financed mostly with debt instruments, such as bonds.
Alternatively two banks could agree to issue capital to each other.
The holding company method is a way of creating capital that is
really debt, and this would show up in consolidated accounts.
Moreover, it is a highly speculative form of financing because
the holding company ends up with financial commitments that must
be serviced out of less liquid assets whose value and income is
much less certain that the holding company's liabilities. This
is why, in the past, bank holding companies were strictly regulated.
The second method gives rise to a "layering" of capital
within the financial system, with banks holding other banks' capital
in their assets. Such cross-holdings have adverse effects on competition
among banks, on bank efficiency, and on the vulnerability of banks
grouped in this way to any risks which might affect the assets
of a single one of them.
III. International bank stabilisation
21. The arguments presented here apply to
a national economy. They apply even more to the international
economy insofar as national financial systems are integrated into
an international system. In a situation in which banks have cross-border
exposures in the form of assets in other countries, exposures
made more risky by the current mix of floating exchange rates
with such rates fixed in particular regions (for example, the
Euro-zone and associated member states of the European Union)
the present Basle Agreement and European Commission Directives
would require even higher capital ratios, than if individual bank
assets and liabilities were confined to only one country. Those
even higher capital ratios imply an even greater indebtedness
22. In such an internationally integrated
financial system, banks and economies would be much more effectively
stabilised if cross-border lending to the private sector were
matched by a commitment to lend, in the domestic currency of the
bank, to the government of the country of that private sector,
in the event of lending to that country being reduced. Thus, if
a bank located, for example, in the U.K. lends to companies in
South Africa, the bank would commit itself to lend to the South
African government the equivalent of any reduction in lending
by the bank to those companies. In this way, capital outflows
would be matched by new capital inflows to governments which would
then be in a position to stabilise the foreign borrowing of banks
and companies in their respective countries.
23. The government bonds thus issued to
foreign banks would be long-term, or at least sufficiently long-term
to avoid repayment pressures on the issuing governments in the
midst of crisis. Banks holding such foreign government bonds should
be given an option to sell them after a given period, but before
they are repaid, to a multilateral monetary agency such as the
International Monetary Fund at a price mutually agreed.
24. Such a system of advance, or contingent
lending commitments would encourage due caution with cross-border
lending, while stabilising cross-border bank capital flows. It
would obviously be least taken up in countries with relatively
little private sector foreign borrowing, such as some of the major
OECD countries. However, smaller countries and emerging, developing
and transition economies, their banks and their companies, could
benefit from the greater stability that such a system would bring.
In this way, contingent commitments to lend to governments in
the event of private sector lending withdrawals would strengthen
and stabilise the international financial system.