The theory of moral hazard has been invoked more recently by the Governor of the Bank of England, Mervyn King, in initially
resisting a bail-out of Northern Rock. His has been a more sophisticated version of the argument than Mellon’s, based less
on self-righteousness and a desire to punish the imprudent than a practical concern that free insurance or underwriting from
the government would encourage further excessive risk-taking. The experience of the Greenspan years was that, if the US Federal
Reserve intervened quickly to cut interest rates drastically at any sign of a potential financial crisis, it would lead to
a new wave of imprudent investment behaviour when the economy recovered. Financiers came to accept such intervention as normal,
and as a duty of government.
This view was put, in parody form, by a leading US hedge-fund
manager, Jim Cramer, who lost his temper on CNBC television when the financial storm broke in August 2007, accusing the Federal
Reserve of being ‘asleep’ and Mr Bernanke of ‘behaving like an academic’, and demanding help for ‘my people’ (that is to say,
Wall Street). Much more abuse of the same kind was directed at Mervyn King in London for not opening his cheque book sooner.
In practice, in the early nineteenth century an approach to financial crises was developed pragmatically, by trial and error,
and was later rationalized by Walter Bagehot. The resulting rule was that it is the job of central banks to advance liquidity
to other banks when required, but only at a penalty rate, against sound collateral, and not to institutions that are insolvent.
A procedure developed about two hundred years ago, and crystallized 130 years ago, has survived remarkably well the big changes
that have subsequently taken place in banking. But there is much scope for misunderstanding over what the rule means in practice,
since suitable collateral is a matter for judgement and the distinction between solvency and illiquidity can be less than
clear.
Financial commentators and financiers unfavourably contrasted the reluctance of the Bank of England to assist banks during
the crisis of August 2008 with the greater willingness of the European Central Bank and the Federal Reserve. The Bank of England
took a less accommodating approach to collateral; it was, understandably, reluctant to accept mortgages on taxpayers’ behalf
in a falling housing market. But there was a more fundamental point. Mervyn King, in a comment that was to create a serious
hostage to fortune, gave a classic statement of the case against indulging moral hazard a few days before the rescue of Northern
Rock: ‘The provision of large liquidity facilities penalizes those financial institutions that sat out the dance, encourages
herd behaviour and increases the intensity of future crises.’ Not only did the Governor then have to acquiesce in the rescue
operation for Northern Rock, but several months later opened a special liquidity facility from which bankers could borrow,
albeit with
a penalty. The European Central Bank, by contrast, appeared to be willing to lend as and when required, without a penalty
rate. And after Mr Cramer’s tantrum was taken to heart, the Federal Reserve was enthusiastically praised by Wall Street. Perhaps
that was because it did what was asked of it, and in its later rescue of Bear Stearns, and then Freddie Mac and Fannie Mae,
went beyond the traditional role of lender of last resort by rescuing companies from threatened insolvency.
The three main central banks affected by the crisis have carried out their classic lender of last resort liquidity functions
with varying enthusiasm and alacrity. There has been less common ground, and greater divergence, in the practical meaning
of moral hazard in respect of rescue operations for failing institutions. As we noted above, when the Bear Stearns operation
took place, the Federal Reserve acted speedily, through a guaranteed line of credit, to ensure that the bank was taken over,
by J P Morgan Chase. The US authorities were less interested in the long-term risks of moral hazard than in the immediate
consequences of bankruptcy triggering widespread default on the banks’ obligations in respect of derivatives. The state partially
stabilized some of the risks of future losses. The shareholders were reprieved; instead of losing their shirts, they were
allowed to retain roughly $1 billion in value. J P Morgan Chase, which took over the bank, had an opportunity to profit from
any recovery, while benefiting from taxpayer guarantees, the full magnitude of which is not clear.
Those who felt queasy about this use of the economic muscle of the state to support supposedly risk-taking, profit-seeking
firms had even more reason to worry about the rescue of Fannie Mae and Freddie Mac. These privately owned bodies were given
limitless state guarantees. No change was demanded in a management team whose business model, reinforced by personal incentives,
had created excessive risk. And there were continued dividends for shareholders, who had already benefited substantially from
earlier implicit government guarantees.
It was only after a time lag of two months and a further bout of
uncertainty in the markets that the US authorities introduced a new set of controls over the private institutions they had
rescued. The US authorities gave ample demonstration of Martin Luther King’s description of his country’s approach to policy
half a century ago: ‘socialism for the rich and rugged free-market capitalism for the poor’. It was primarily a belated anxiety
about moral hazard that then persuaded the US authorities to let Lehman Brothers go bankrupt, rather than to rescue it like
Bear Stearns. Yet the later Paulson plan was full of moral hazard – taxpayers offering to take over the bad debts of the most
irresponsible banks.
The alternative approach to rescue involves securing gains for the taxpayer, and avoiding moral hazard, by nationalizing failing
institutions, replacing their management and then selling them on in improved economic conditions, without rewards for the
investors whose institutions had failed. The US had employed this approach in the past, with the Continental Illinois Bank
in 1984. More systematically, it was used by Sweden and other Scandinavian authorities in the early 1990s, as we have already
noted. There was a major banking crisis costing the economy 6 per cent of GDP between 1990 and 1993, which was dealt with
by a mixture of bank closures, government-sponsored reconstruction, and temporary nationalization under the direction of a
Bank Support Authority.
Britain has finished up in a similar place to the Swedes. It first struggled with the problem of Northern Rock, nationalizing
it, but only after months of indecision. The initial hope was a Bear Stearns-type rescue by Lloyds, involving a £25 billion
government guarantee. It was never clear, however, what the terms of such a deal were and particularly how the risks and losses,
or potential profits, were to be allocated between the government and the private sector. For several months the government
sought, in the full glare of publicity, to effect another private sale, to Richard Branson or other potential buyers. But
the same set of problems proved insurmountable: how to ensure that the risks to the government of continued loans and guarantees
would be properly
offset by appropriate rewards; that the new private owners would share properly in the risks; and that current shareholders
would not profit from government guarantees. The government also indulged for far too long the odd notion that the Northern
Rock management had made no mistakes and were therefore part of the solution. When the banking crisis struck with full intensity
in the autumn of 2008 lessons had been learned, and when it was decided that Bradford & Bingley had to be rescued, it was
nationalized promptly. On a totally different scale, the Royal Bank of Scotland has been taken in to public ownership without
being fully nationalized.
Some lessons have been learned from these contrasting experiences about how to reconcile rescue operations designed to preserve
financial stability with the avoidance of moral hazard. Nationalization is one route, though there are other ways of striking
a proper balance, through strict conditions for rescues: upper limits on assistance; drastic reorganization; removal and,
if necessary, punishment of existing management; penalty rates on credit; a freeze on dividends during the lifetime of a rescue;
guarantees for the government of participating in the potential upside. The British recapitalization plan for RBS and Lloyds
was of this kind though the conditions turn out to have been weak. The rescue victim might baulk at such conditions – and
they have – in which case the government has the option of full nationalization. Seen in this way, far from temporary nationalization
being a step towards socialism, it is an essential tool for managing a market economy and maintaining its disciplines in a
financial crisis.
Questions of moral hazard do not stop at institutions and shareholders. They also affect depositors, the millions of individuals
whose savings form the basis of the banking system. The dilemma is this: if depositors fear that they might lose their money
if they leave it in a bank, they will incline towards safer but less productive options, such as hiding it under the bed,
buying gold and jewellery or land, or spending it. But if they are fully protected from the risk of losing their money they
may flock to banks that
offer higher returns by cutting corners and taking excessive risks. There is a tricky balance to be struck. The trickiness
is made more difficult by the fact that banking is inherently risky and rests ultimately on the hope that depositors will
not all ask for their money back at the same time, since most of it is tied up in illiquid assets.
The key turning point in depositor protection was in the 1930s. Prior to that banks, by and large, depended on their reputations.
Financially conservative banks attracted depositors through periods of financial turbulence precisely because they were, or
claimed to be, very cautious in their use of savings, investing heavily in government paper or businesses with good collateral.
In the UK, reputation has also been the mainstay of a system that rested, at least for the last century, on a small number
of large banks that had never seen a run by their depositors. The USA, however, had a somewhat more freewheeling system in
which banks occasionally sank and savers drowned with them. The Great Crash led to the biggest bank run in history and in
turn to the establishment of deposit guarantees operated by the Federal Deposit Insurance Corporation. These initially covered
deposits of $10,000, which was later raised to $100,000. The FDIC financed its operations by collecting premiums from banks,
which passed on the cost to their customers. The FDIC had plenty of practice, mostly with tiny banks, and it worked well in
stopping runs. It ran into two difficulties, however. One was that when a really big bank failed – like Continental Illinois
in 1984 and First Republic Bank of Dallas in 1988 – it felt obliged to abandon the upper limit in order to prevent panic.
The other was that its purposes (and those of sister institutions) became subverted by their being given a central role in
rescuing banks, as opposed to their depositors, using taxpayers’ money.
By contrast, European post-war banking systems have been tightly controlled and, in some countries, nationalized, partly in
response to the banking disasters of the 1930s. So issues of depositor protection have been less in evidence than in the USA.
In the
UK there was the additional protection of informal guarantees between the banks, which were formalized in 1890 when Barings
Bank capsized and others, such as Martins, were threatened. There was also a post-war system of depositor protection – insuring
deposits up to £35,000 – but it was hardly used, largely because the high-street banks were assumed to be totally reliable
(and protected, as institutions, by the mysterious but seemingly definitive lender of last resort role of the authorities).
Those portentous bank branches that dominated the high street, and which used to be inhabited by smartly dressed, socially
superior staff, overseen by a terrifying manager, were the embodiment of reliability. It was a privilege to be allowed to
bank there and, even more, to borrow. Depositors could sleep safely knowing that improvident riff-raff were being kept at
bay and that the army of bean-counters knew how to add up.
Mrs Thatcher’s financial reforms of the 1980s radically changed the high-street banks from being safe but boring to being
more aggressively competitive – but also, we have discovered, less secure. In particular, they competed to offer loans on
attractive terms. The credit card revolution further liberalized lending. The proliferation of banks in the 1990s, with the
demutualization of UK building societies, added to competition. Perhaps someone in the Bank of England or the Treasury should
have stopped to think about ‘what if’ scenarios, such as the risk of a small but ambitious bank behaving recklessly and putting
its depositors at risk. But no one did. Until Northern Rock. It soon became clear that very few of Northern Rock’s depositors
knew that they were protected or, if they did know, did not trust the system to pay up (they were right, in the sense that
the process is cumbersome and takes months). They panicked, and Britain suffered the first bank run for over a century. It
was only stopped by the Chancellor offering unlimited guarantees to depositors (as the US authorities had done in the 1980s
to head off the run on Continental Illinois).
A sense of panic resurfaced in the middle of the September– October 2008 banking crisis when Ireland sought to prevent a
run on its banks by offering unconditional guarantees, leading Greece and some other European countries to follow suit. The
UK came close to being forced to follow, but a combination of the reassurance (for depositors) of nationalization as a last
resort and a depositor protection scheme (now being improved by parliament) prevented further panic. Even private depositors
in the risky but high yielding Icelandic banks were fully protected (but not councils or charities).