The following speech was presented by Professor Forrest Capie at the CMRE meeting, "A Time of Financial Fragility and Latent Instability* Waiting for Solon?" held May 10, 2007.

The search for stable prices and the role of gold

Introduction

How can stable prices be secured so that the economy can be allowed to deliver what it is capable of? There was a time when the question simply did not arise.  Money was tied to gold and stability prevailed. At the present time some monetary economists have moved to the other end of the spectrum, moved so far that not only is metal an irrelevance but money itself is no longer of importance.  History has lessons for those who suggest dispensing with monetary indicators, history would say, ‘not so fast’. 

There is wide agreement that price stability is highly desirable.  Further, that there is no trade-off between inflation and output in anything other than the very short run, and even that is not clear.  Money growth in excess of the growth of output will eventually find its way into prices though there may be long and variable lags. I shall present briefly three kinds of historical evidence in support of the importance of money for stable prices. The first relates to the very long run.  The second takes some short-period extreme examples of inflation from history, and the third reminds us of the 1970s and what happens when the cause of inflation is forgotten. 

The search for stability 

The search for efficient money – something that holds its value (in other words provides stable prices) -  is something that has gone on over millennia.  Given that the features of money are that it should act as store of value, a unit of account, and a medium of exchange and that the characteristics necessary for these are durability, transportability, and acceptability, it is no surprise that precious metals, and in particular gold and silver, should emerge as they did to perform their useful function a long time ago.  Goldsmiths played an important part in the origins of modern fractional reserve banking systems for they soon learned how to hold certain proportions of the gold deposited with them and to lend on the strength of the balance.  The bank note was a short step away and can be found in the seventeenth century, based of course on metallic backing. Each step in the development of money was a way of economising without sacrificing confidence and credibility.  The move from barter to primitive money (e.g. cowrie shells) removed the need to find a coincidence of wants. The move from primitive money to metals had further economies, as did the move from gold and silver to paper based on them. However, the step that moves from that to paper not based on any tangible backing severely tests credibility and on occasions breaks down altogether.  (It was when America abandoned metallic backing in the Civil War that the US dollar bill first carried on the back the phrase, ‘In God we trust’.)  Be that as it may, that has been the road travelled and the question that arises is how robust are non-backed monetary systems that we now have, and have universally for the first time in 2,500 years.  Have we reached a plateau of such advanced understanding that we can now proceed happily with fiat money? 

Charles de Gaulle once said, “There can be no other criterion, no other standard than gold.  Yes, gold which never changes, which can be shaped into ingots, bars, coins, which has no nationality and which is externally and universally accepted as the unalterable fiduciary value par excellence.”  De Gaulle thereby summarised a long-held view of gold and the many reasons for the view being held.  It is durable, divisible, and for many years over most of the world was the ultimate standard of value. It was a standard that held its purchasing power in terms of goods over long periods.  For close to two centuries gold was the basis of the world monetary system.  Whether gold is money, or the basis of the monetary system, means linking a currency to gold at a fixed price. The behaviour of prices was therefore taken outside the control of governments, and depended rather on gold supply relative to the demand for it.  This provided an automatic stabilising mechanism. Price performance was not accidental.  If prices were falling because the supply of gold was short of the demand for it, there was an incentive to produce more gold and vice versa. The price level would fluctuate about a flat trend.  There developed an expectation that prices would be stable.

The classical gold standard reached its high point in the late nineteenth century. The period is usually dated as 1880 – 1914 and was a period of great prosperity and of considerable stability, and indeed was the first period of globalisation. Most countries joined the gold standard or aspired to join it and so behaved in a way that was consistent with membership. Prices fluctuated a little from year to year but over the long term were stable and flat.  Prices in 1914 were broadly where they had been in 1870. The gold standard was seen as the optimal arrangement not least because it left the system immune from political interference.  There were enormous capital flows around the world as people in one country willingly invested in others often via long-term bonds, so facilitating long-term development around the world.  

The long run

When money is under such control stable prices follow. A history of world prices from the beginning of time until the nineteenth century can be covered very quickly.  There is not much to say. Prices were essentially flat for century after century. There was no such thing as inflation over long periods.  There may be some cyclicality and even longish cycles on occasions but prices tended to return to where they had been in the past.   There is talk of the Roman Empire being destroyed by inflation but in fact the best estimates show that the annual rate of inflation across the fourth century AD was in very low single figures.  The scope for monetary expansion in the absence of paper money was limited. The same is true for the period in the sixteenth and seventeenth centuries in Europe when there was a large influx of precious metals from the new world.  In England the rate of inflation in the sixteenth century was between 2 and 3 per cent.  Again, while the money stock increased, the scope for increase was limited. From the middle of the seventeenth century until the end of the eighteenth century the world at large was in a period of flat prices.

There was an interlude during, and for a few years after, the Napoleonic Wars when Britain suspended convertibility into gold – from 1797 to 1821.  The threat of invasion by France became real and it was rumoured in 1797 that it had taken place. A crisis blew up and the standard was suspended.  Inflation then took hold and averaged around 2 – 3 per cent per annum across the period, and the exchange rate depreciated. A great debate also blew up over causation – was it real or monetary factors that caused the inflation? This is the beginning of modern monetary economics and this particular debate was won at the time by the ‘bullionists’ who believed in the need to restore the gold standard. And with the standard restored again in the long nineteenth century (1820-1914) prices in England were stable, at the end being close to those obtaining at the beginning.  The explanation for this stability can be found in the metallic base of the monetary system that anchored prices and price expectations firmly. Large increases in the money stock awaited twentieth century technology.  

Serious episodes

So across the long run and around the world prices were stable when they were linked to precious metals. When, however, metallic backing was abandoned there were episodes of serious inflation. There were just a few experiences of what might be called very rapid inflation – inflation in excess of 100 per cent per annum – prior to the twentieth century. Then in the first half of the twentieth century there were several clear cases of hyperinflation, something much worse. The second half of the twentieth century is replete with examples of inflation at stratospheric heights.    Between 1960 and 1992 world prices rose seventeen fold.  In Latin America they rose 14,000,000 fold.  In individual countries they rose by much more. In August 1993, for example, inflation in Serbia was running at an annual rate of 360 million billion per cent if that is an expression that can have much meaning. Anyway, a lot.

But before the twentieth century there were only a very few. One was the American War of Independence; one the French Revolution; and a third the American Civil War.  I think it is true to say there were no others. (There are some suggestions of a similar experience in China but data difficulties prevent discussion.)  In the 1920s there were five cases of hyperinflation: in Russia, Hungary, Austria, Germany, and Poland.  In Germany, the worst experience, prices rose 1,000 million fold across 1923/24.  In the 1940s there were three such examples in China, Greece, and Hungary again. The last was the worst ever up to that time with numbers too big to comprehend.  Stories circulate of how in some department stores in Budapest a bell would ring every so often to indicate that as of that moment prices had doubled.

In all the experiences of such inflation it was with an unbacked paper currency – of the kind we now have everywhere.  A vast expansion of paper currency preceded or accompanied all these inflations.

What is also common to all these inflations is that there were large and growing fiscal deficits, and deficits like these require monetizing which is what happened in these cases. Further, what all the examples have in common is not that they were found during wartime but rather that it was civil war or at least serious social unrest that gave rise to the inflation. The external threat is not critical because it usually stimulates patriotism, and tax revenue can be raised and borrowing can be carried out more easily. When the threat is internal there is a loss of tax revenue, a diminished ability to borrow, and a temptation to buy off the opposition or find the resources to quell the rebellious faction.  A cursory glance at many of the experiences of very rapid and occasionally hyperinflation in the second half of the twentieth century confirm this story:  from Indonesia in the mid-1960s through much of Latin America, and Africa, to the Balkans in the 1990s and to Zimbabwe at the present time. It is civil war or serious social unrest that prompts governments to turn to the printing press and buy resources.

Governments have always been keen to get hold of resources and in times of great difficulty the temptation is to confiscate them. When tax revenues and borrowing are not possible the easiest way of doing this is through the inflation tax.  As Keynes put it, inflation ‘is the form of taxation which the public find hardest to evade and even the weakest government can enforce when it can enforce nothing else’. 

So what the extreme cases of inflation before 1950 suggest is that weak governments (weak measured in terms of the opposition to them) who cannot tax or borrow sufficiently turn to the printing press and generate inflation.  And the more extreme examples of the second half of the twentieth century have the same characteristics. 

1970s

Thirdly, the 1970s. After the Second World War the prevailing economic wisdom was that money did not matter.  It was claimed that there were superior ways of organising our affairs, and that the domestic economy could be managed in such a way as to deliver full employment and economic growth. A pegged exchange rate could look after the international dimension. Under the accepted accounts inflation was not after all a monetary phenomenon but a consequence of cost push pressures.  In fact interest rates were sometimes seen as a cost in this story and raising interest rates was therefore seen as a factor causing inflation!  But the chief cost was wages and therefore the solution was to be found in a prices and wages policy, often seen as needing the co-operation of the trades unions.

Inflation picked up in the 1950s and was accelerating in the 1960s and incomes policies were applied and relaxed throughout the period.  But they did not work.  In the 1970s inflation was rampant reaching double figures in many OECD countries, and peaking in Britain at about 30 per cent in the mid-1970s.  This was the worst rate of any time in Britain be it war or peace.  Not only was the inflation the worst of all but output performance was much worse than it had been and unemployment at its highest, stagflation as it was dubbed at the time.

A large body of academic work was beginning to build in the 1960s that showed that money mattered after all and slowly this seeped into the thinking of the monetary authorities and ultimately into policy. Gradually, though understandably for people who had invested so much intellectual capital in the opposite case, not enthusiastically, it came to be accepted that excess money was after all the cause of inflation and a policy of monetary targets was pursued.  In Britain this was not done with any great thoroughness in the 1970s but it was a step in the right direction. When monetary policy was more diligently employed inflation was brought down.

There came to be a fairly widespread acceptance of these views and the techniques for achieving success in practice.  There was also some recognition that the policies were most likely to be successfully implemented by independent central banks.  Central bankers were commonly more conservative in their views on monetary policy than governments.  But they frequently complained that governments made their jobs impossible by overspending and obliging them to finance some of the fiscal deficit.  So independence alone is not a panacea.  Governments still need to live within their means. 

Conclusion

Let me conclude with a brief summary.  The gold standard in its classical form delivered price stability and allowed the economy to do whatever it could with uncertainty reduced to a minimum.  What was equally important was that there was internal price flexibility.  More importantly, the government having accepted the gold standard rule was bound to behave properly fiscally.  If it did not the country would be forced to abandon the standard, lose credibility and find it expensive to borrow.

The use of gold was highly successful in providing a stable price level.  But there came time when there was not enough to go round – not enough to provide the reserves for a rapidly growing world economy. New solutions had to be found. Furthermore, when internal price flexibility is lost as it is with regulation - minimum wages, welfare payments and so on – flexible exchange rates might be desirable since it gives a greater degree of freedom in policy. But that regime needs to be accompanied by monetary and fiscal prudence.  There still needs to be a commitment to fiscal and monetary probity. 

Finally, I come back to a question posed at the beginning. Are we now sufficiently advanced to operate with unbacked paper money - fiat money? A fear must surely be that fiat money will once again go off the rails.  In the late eighteenth century the French had a strong aversion to paper money deriving from an earlier experience The Revolution brought a change in attitudes and the great statesman Mirabeau who had once called paper money a ‘loan to an armed robber’ now felt that the French had become more enlightened and that assignats could be created to an amount sufficient to cover the national debt.  Prices rose from a base of 100 in 1790 to 38,850 in May 1796.

Currently, comparative price stability seems to prevail. It is difficult to be sure how much credit should go to the central banks since there has been a benign environment in terms of international pressures.  This line should not be pushed too far however, for you can have whatever level of inflation you choose, high or low.  But in a period when the desire has been for low inflation the authorities have at least been helped rather than hindered by the nature of the pressures.  Currently, inflation targeting is popular.  It is an improvement on monetary targeting if it can reduce the cost of downturns in the economy. But it does allow quite considerable discretion in monetary policy, as against operating under a rule such as the gold standard, and the use of discretion has come unstuck in the past.  The great danger is that if inflation did edge up and inflation expectations were to change, it may be difficult to contain the new inflation and take some time to alter expectations. Surely it is better then, at the very least, to keep an eye on the monetary aggregates and be prepared to see them as a useful indicator of underlying inflationary pressures.