Each separate panic has had its own distinctive features, but all have resembled each other in occurring immediately after a period of apparent prosperity, the hollowness of which it has exposed. So uniform is this sequence, that whenever we find ourselves under circumstances that enable the acquisition of rapid fortunes, otherwise than by the road of plodding industry, we may almost be justified in auguring that the time for panic is at hand."

That could have been said yesterday. In fact it was written in 1859, in a history of the commercial crisis of 1857-58. Financial shocks all feel different, but most of them are pretty much the same.

Let's start with an obvious parallel. The run on Northern Rock was the first big bank failure in this country since 1866, when Overend, Gurney went down. Both were long-established institutions with histories of prudent and highly respectable finance: Northern Rock as a mutually owned building society in the north-east, Overend with its roots deep in Quaker East Anglia.

The leadership of both had been taken over by a new breed of growth-hungry executives, anxious to expand their loan books rapidly: Northern Rock in mortgages at the top of a house-price bubble, Overend in some decidedly dubious enterprises including shipbuilding, grain trading, railway finance and much else besides. Both made the fatal mistake of relying on short-term borrowing to fund their rapidly expanding and increasingly risky loan books. And just like Northern Rock, Overend paid the price for its flawed business model. The one big difference was that the Bank of England let Overend fail, and take down other firms with it in the ensuing panic.

Two American economists, Carmen Reinhart of Maryland and Ken Rogoff of Harvard, have recently published an analysis of the current financial crisis in the context of what they identify as the previous 18 banking crises in industrial countries since the second world war. They find what they call "stunning qualitative and quantitative parallels across a number of standard financial crisis indicators" - the common themes that translate these individual dramas into the big-picture story of financial boom and bust. Their study is focused on the US, but if you look at the relevant numbers, most of their analysis also applies to the UK.

Ahead of each big financial shock, house prices rose rapidly, as did equity prices. Current account deficits ballooned, with capital inflows accelerating up to the eve of the crisis. Rising public debt is a near universal precursor of other postwar crises. And overall economic growth started to fall away as trouble loomed. In addition, financial shocks in the past have often been preceded by periods of financial deregulation.

I witnessed as a wide-eyed reporter the secondary banking crisis of 1973-74, which came within a whisker of pulling down some of our most powerful financial institutions; and I still have in my possession the press release put out by one of the big four clearing banks denying that it was in difficulties. That drama had several different triggers: one was the Competition and Credit Control scheme introduced at the end of 1971, which removed the ceiling on loans, reduced banks' liquidity requirements and ended the interest rate cartel. Over the next two years, total credit advances to UK residents multiplied 2.5 times.

Deregulation has also played a part in today's events. To quote Paul Volcker, former chairman of the US Federal Reserve, over the past 25 years "we have moved from a commercial-bank-centred, highly regulated financial system, to an enormously more complicated and highly engineered system. Today, much of the financial intermediation takes place in markets beyond effective official oversight and supervision, all enveloped in unknown trillions of derivative instruments."

Another common feature, this one identified in a recent FT article by Charles Goodhart and Avinash Persaud, is that asset-price bubbles often follow periods of price stability. They cite as examples the US in 1929, Japan in the 1990s, Asia in 1997-98 and the implosion of subprime mortgages in 2007-08.

Low inflation leads to low interest rates and the accumulation of real assets. That in turn leads investors to take bigger risks in order to secure a higher return on their investments. A classic example came in the 1970s, when petrodollars flooded the international capital market and drove interest rates down. Banks were flush with liquidity and looking around for new outlets. They found what they were looking for in the developing economies of central Europe and Latin America. But things came to a grinding halt in August 1982, when the Mexican government suspended debt service.

Much the same has happened in the past few years. Indeed Rogoff and Reinhart rather cheekily suggest that this time round, a large chunk of money has effectively been recycled to a developing economy - but one that this time exists within America's own borders. More than a trillion dollars were channelled into the US subprime mortgage market, which is made up of the poorest and least creditworthy borrowers.

The common feature here is that risk is misunderstood, and so mispriced. In 1929, it was excess leverage in investment funds that went wrong. In the Asian crisis of 1997-98, it was about mismatched currency exposures. In the dotcom boom, the thought was that the internet would lead to everlasting growth. And in the recent credit bubble, the idea was that by slicing and dicing debt, and distributing it widely around the world, junk debt could be miraculously converted into triple-A investments.

Is it really any different this time?

Yet another feature that this latest drama has in common with other similar episodes in history is that it is truly international in character. Time and again over the centuries it has been clear that optimism, greed, euphoria and despair do not recognise national boundaries. Well before the internet, changing market moods swept across the world with astonishing speed.

The falls in share prices on October 24 and 29 1929, and again on October 19 1987, were practically instantaneous in all financial markets, except Japan. This was far faster than could be explained by arbitrage, capital flows or money movements. The South Sea and Mississippi bubbles of 1720 were related, stoked by deregulation and powerful monetary expansion in England and France. And the crisis that followed them rippled across the Netherlands and northern Italy, as well as northern Germany. The list of such international earthquakes is just about endless.

Of course there have always been physical connections between national economies around the world - internationally traded commodities and bullion, exports and imports, capital and money flows, and so on. But what I find fascinating are the purely psychological connections, as when the mood of investors in one country infects those in another, sometimes great distances away. And what we always have to remember when thinking about financial euphoria and panic is that rational behaviour can very often go out of the window. Isaac Newton, one of the greatest minds in history, speculated wildly in South Sea stock and ended up losing his shirt. As he observed glumly: "I can calculate the motions of the heavenly bodies, but not the madness of people."

What are the consequences of big financial shocks?

The great source of knowledge on this is the economic historian charles kindleberger, whose classic book, Manias, Panics and Crashes, was published in 1978. Kindleberger's view, rather tentatively expressed, is that the role of lender of last resort, properly exercised, is the key to shortening the business slowdowns that normally follow financial crises. He cites as evidence the crashes of 1720, 1873, 1882, 1890, 1921 and 1929. In none of these was a lender of last resort effectively present. The depressions that followed them were much longer and deeper than others. Those of the 1870s and the 1930s were both known as "the Great Depression".

The role of lender of last resort was classically defined by Walter Bagehot. His great book Lombard Street was published in 1873, and set out what has become the guiding mantra for central banks in times of crisis ever since: lend freely at high rates against good collateral. Lend freely, in his words, "to stay the panic". At high rates, so that "no one may borrow out of idle precaution without paying well for it". And lend on all good banking securities to an unlimited extent - because the "way to cause alarm is to refuse someone who has good security to offer".

But, unfortunately, life is not as simple as that. For one thing, central bankers don't always do the job well. The Bank of England is generally thought to have played a poor hand in its intervention in the panic of 1825. City historian David Kynaston shows how its policy veered wildly between complacency and an over-sharp contraction of credit. More than 70 banks collapsed, and according to legend the Bank of England itself narrowly escaped disaster. Just as it ran out of £5 and £10 notes, someone discovered a block of £1 notes left in the vaults since 1797. These were issued with government approval, and "worked wonders". The ensuing depression lasted several years: by the end of 1827, according to one report, "in commerce, almost every one still smarting under the losses which the climax of 1825 had left them - and fearing from the long continuance of the swell after the storm even now to venture far from shore".

Another problem is that over-enthusiastic central bank intervention can store up real trouble for the future. This lies behind the energetic debate over the economic legacy of Alan Greenspan, who stepped down as chairman of the Federal Reserve in January 2006. The classic role of the Federal Reserve has been to lean against the wind, easing credit in hard times and tightening it before things get out of hand. In the words of William McChesney Martin, who served 18 years as chairman, up to the time of President Nixon: "The function of the Federal Reserve is to take away the punch bowl just as the party is getting good." Greenspan's critics claim that far from taking away the punch bowl, he was only too happy to chuck in an extra bottle of brandy at the first sign of the party coming to an end.

They also say that no matter what went wrong, the Fed under chairman Greenspan would save the day by creating enough cheap money to buy off trouble. After the crash of October 1987, the Fed cut rates three times in six weeks - and stocks quickly recovered. The same happened after the Asian crisis in 1997-98, and again after 9/11.

But there are limits to how far central banks can go. Cheap money and negative real interest rates lead over time to frothy speculation and inflation. This is what central bankers mean when they warn about the dangers of moral hazard - the risk that bailing bankers out of trouble will only encourage them to be even more irresponsible in future. The great difficulty lies in attempting to draw a distinction between individual culpability, when you should let an institution go bust, and the risk of systemic failure that might have desperately serious consequences.

For this reason, banking authorities over the years have often resolved not to intervene, only to find themselves forced to cave in under pressure. Lord Liverpool threatened to resign as chancellor of the exchequer in 1825 if the speculators were bailed out - but eventually they were. Kindleberger finds similar examples in 1763, 1869, 1897 and 1975, with the rescue of New York City. And Mervyn King, the present Bank of England governor, also made powerful warnings against the risks of moral hazard until the Bank was ready to make what looks like a classic Bagehot-style intervention a couple of months ago.

It's a difficult balance between teaching speculators a lesson and averting systemic failure. Only history will be able to judge the different approaches of Greenspan and King.

In its response to the latest crisis, the US Federal Reserve has gone considerably further than its predecessors. Faced with the threatened collapse of the Wall Street investment bank Bear Stearns, with its huge exposure to the derivative and securitised loans markets, the Fed, to use Paul Volcker's words again, judged it necessary to take actions that "extend to the very edge of its lawful and implied powers, transcending certain long-embedded central banking principles and practices".

As such, this bail-out must surely lead to a radical reshaping of US securities regulation. If the Fed is to stand behind such institutions in future, it will require a much more direct degree of supervision than permitted under current rules. No wonder Volcker sounds concerned.

How long and deep is any economic slowdown likely to be?

Of course any answer has to be hedged around with lots of variables. The key one is the presence and performance of a lender of last resort. But there are many others, because booms and busts seldom have a single trigger. For example, the crisis of 1847 had the railway mania, the potato disease, a wheat crop failure one year and a bumper crop the next, followed on the continent by revolution. The Wall Street Panic of 1857 was turned into a prolonged recession by civil war. Libraries of textbooks have been written on the causes of the 1930s depression, to which the Great Crash was by no means the only contributor.

The economic challenges today are made all the more daunting by the fact that commodity prices, particularly energy, are soaring at precisely the moment that a very large hole has been blown into the balance sheets of the world's most important banks - another case where it's not just the financial shock that is causing the trouble. Rising inflation plus slowing demand add up to serious problems for policymakers everywhere. Thus the European Central Bank recently raised interest rates to keep inflation in check, at the very time when slowing demand would normally require a cut.

In spite of all these variables, history does provide some pointers to the potential economic impact of a big financial drama. As analysed by Reinhart and Rogoff, in the five most catastrophic examples since the second world war (in Finland, Japan, Norway, Sweden and Spain), the drop in annual output growth from peak to trough was more than 5 per cent and growth remained well below pre-crisis trend even after three years. Taking all 18 shocks together, the average drop in real per capita output growth turned out to be a bit more than 2 per cent, and it typically took two years to return to trend.

Economist Paul Ormerod recently analysed 255 examples of recessions in 17 western countries between 1871 and 2006. He found that 164 of them lasted just one year, and the great majority were over within two years.

As a very rough rule of thumb, then, up to about two years of slowdown following a financial shock has not been out of the ordinary over the years - provided the lender of last resort has done its stuff. On that basis, and given the sheer scale of the current financial dramas, the US and UK Treasuries always looked rather optimistic in their projections of a short, sharp setback after the summer of 2007, with recovery beginning to make itself felt in the second half of this year. As things stand, the outlook for 2009 looks likely to be rather worse than that for 2008, at least in the UK.

How much do financial traumas lead to economic slowdowns?

First, they lead to a massive destruction of assets and wealth, with all that implies for economic activity - just consider how much poorer shareholders in the world's big banks feel today compared with a year ago. The impact of the shock on bank balance sheets also has an immediate effect on their ability to lend. The most spectacular examples of severe credit constraints came in the Great Depression. Automobile sales in America fell from 4.5 million in 1929 to 1.1 million in 1932, and didn't climb above their previous peak for 20 years.

In addition, the swing from boom to bust completely changes the public view of finance, and the willingness to take risks. Yesterday's heroes become today's villains. Investment horizons shorten. As J.K. Galbraith observed in his magnificent book The Great Crash 1929, within a few days of the downturn "something close to universal trust turned into something akin to universal suspicion".

Financial shocks also frequently expose fraud and crime, adding to the general sense of unease and leaving the public baying for vengeance. The directors of the South Sea Company barely escaped with their lives. One parliamentarian argued that they should be found guilty of parricide and suffer the ancient Roman punishment for that crime - to be sewn into sacks along with a monkey and a snake, and drowned. One thing to watch out for when things are going wrong is employees who never take a holiday. In London, it was only when William Pullinger of the Union Bank was obliged to attend a funeral in 1860 that his massive embezzlement came to light. In Paris, it seems that rogue trader Jerome Kerviel, who recently punched a hole in Societe Generale's balance sheet, was similarly disinclined to leave his desk.

Legislators do not always make their wisest calls at such times of financial frenzy. The Bubble Act of 1720 banned the issue of all stocks that were not authorised by royal charter and as such made it difficult to start a legitimate business in Britain for more than a century until its eventual repeal. The Wall Street crash of 1929 led to the Smoot Hawley tariff legislation of 1930, with its devastating impact on international trade. The dotcom bubble in the early years of this century was followed in the US by the Sarbanes-Oxley legislation, with expensive consequences for all US listed companies.

As Martin Wolf has observed in this newspaper: "Everybody involved - borrowers, lenders and regulators, too - are all too often swept away in tides of euphoria and panic. To err is human. That is one of the reasons regulation is rarely countercylical: regulators are swept away as well." So the record suggests that we should be cautious in our regulatory response to the events of the past year. But it does not suggest that we should do nothing.

What should be done?

One suggestion is that we should establish some institution, maybe a refocused international monetary fund, and give it the job of looking for trouble on the horizon. It sounds a fine idea. But history is full of authoritative warnings of trouble to come that have been completely ignored in the euphoria of the moment. Share prices slipped briefly after Alan Greenspan spoke of the dangers of "irrational exuberance" at the end of 1996. But then they raced ahead for most of the next 10 years. Bank of England governor Mervyn King was talking about mispriced credit risks and tougher times ahead well before last summer. But when the market is going up, no one cares about the doomsayers.

That's the argument against the widespread view today that monetary policymakers should take asset price inflation as well as monetary inflation into account when setting interest rates. I was on the Monetary Policy Committee when house prices were roaring away in the three years to 2006, and I can remember thinking: "If our mission was to check house-price inflation (which of course it was not), where would interest rates have to go?" It was very hard to guess at what constituted the sustainable price of housing over the long term. Would we be prepared to knock industry for six simply because we felt that house prices were getting a bit too sporty? I don't think so.

A more interesting idea, advanced most recently by Goodhart and Persaud, is for a regulatory framework that raises bank capital requirements by a ratio linked to the growth of the value of bank assets. The purpose would be to moderate excessive lending, and build up bank reserves during boom times.

Some regulation is inevitable, and the Bear Stearns rescue makes reform essential in the US. In the UK, the government will introduce legislation later this year to tackle some of the shortcomings that became obvious during the Northern Rock affair. Other, more microeconomic changes are also likely, in the workings of the credit ratings agencies, for example. There will, as always, be endless and ultimately fruitless debate about the need to curb City bonuses, about the need for more disclosure and more transparency. The annual reports of the big commercial banks will continue to grow by 20 to 50 pages a year.

There's an obvious need to do more to align the financial interests of employees and shareholders in the banking system, to reduce the incentives for traders to take very large short-term risks. But the likelihood is that for the next few years, anyway, all will be caution and sobriety. Banks will be much more cautious about how and where they place their money. The grey will rule the earth and flashy exuberance will be regarded as, well, vulgar.

That is perhaps the final lesson of history, and no one put it better than Bagehot: "In England, after a great calamity, everybody is suspicious of everybody. As soon as that calamity is forgotten, everybody again confides in everybody."

Richard Lambert is the director general of the CBI and a former editor of the Financial Times. This is an edited version of a lecture given at Templeton College, Oxford.

1866 Overend, Gurney

The bankruptcy of Overend, Gurney & Co prompted Britain's most notorious bank run. The bank's troubles began when it expanded from its core business (trading bills of exchange) and moved into riskier investments such as shipyards. When several creditors collapsed, the bank's shares plummeted. Walter Bagehot called its policies "so reckless one would think a child would have lent better". The day after the Bank of England declined support, Overend suspended cash payments and the run turned into a riot - with 10 further banks suspending payments, 200 companies defaulting and "throngs heaving and tumbling about Lombard Street".

2007 Northern Rock

On September 12 last year, as the credit crunch tightened, Northern Rock was forced to seek liquidity support from the Bank of England in the Bank's role as lender of last resort. News of the loan prompted a run on the bank. Over the next few months, the share price, which had been at an all-time high of £12.14 in February 2007, plummeted to 90p as a host of rescue packages failed. The bank, which owed an estimated £25bn to the Bank of England, was nationalised by the government on February 17.

1720 South Sea Bubble

The South Sea Company won a monopoly on trade with South America in 1711 on the proviso that it assume the national debt caused by the War of Spanish Succession. But Spain's control of the seas left the company little scope for trade, even before hostilities resumed in 1718. However, with a royal charter and directors circulating stories of fictional foreign trading success, a bubble sent the company's share prices soaring from £128 to £1,050 in the first six months of 1720. When the bubble burst, prices plunged back to £175, ruining thousands.

1997 Asian Crisis

After decades of robust economic growth that had earned the region its "Asian tiger" moniker and had seen fast flows of capital into liberalised markets from the developed world, Thailand's decision to float the baht in July 1997 sparked a financial crisis that soon spread to other economies in the region. Soon, the "Asian contagion" crisis went global, taking Brazil and Russia with it.

1929 Wall Street crash

At the turn of the 20th century stock market speculation was restricted to professionals, but the 1920s saw millions of "ordinary Americans" investing in the New York Stock Exchange. By August 1929, brokers had lent small investors more than two-thirds of the face value of the stocks they were buying on margin - more than $8.5bn was out on loan. By October, the situation had become unsustainable and stockholders panicked. In a single day, more than 12 million shares were traded, as people desperately tried to realise their assets. The Great Depression had begun.

1987 Black Monday

On October 19 1987, world stock markets experienced their biggest one-day falls in history (the Dow Jones Industrial Average dropped 22.6 per cent, while the FTSE 100 dropped 10.8 per cent). The precise causes of the crash remain unclear - some have even blamed FTSE's woes on the London hurricane of October 16.