How do we measure wealth? In some unit of account. Money is meant to function, among other things, as a unit of account, the denominator for any given asset value. But if the purchasing power of a currency is unstable, either due to inflation or deflation, this distorts the way in which real wealth is measured. For example, if your benchmark unit of account is the Zimbabwe dollar, you have made outrageous profits during the past few years almost regardless of your choice of investment.
One option for dealing with an unstable unit of account is to measure investment performance or wealth in inflation-adjusted terms. For example, at a given starting point in time, wealth could be indexed to 100 and then, say, an annual return of 10% would take the index to 110. If, however, inflation was 3% that year, then the index would be adjusted lower, to 107%, implying a real return of 7%. While it sounds nice on paper, measuring performance in this way is in fact highly problematic. For example, what inflation index should be used? Are existing inflation indices adequate measures of changes in real purchasing power? And is the cost of living in one country an adequate measure from the perspective of a global investor? The dollar may be the world’s primary reserve currency, but if US inflation is only 4% and inflation in India is 20%, should Indian investors benchmark themselves against US or domestic Indian inflation? Or some weighted combination of the two perhaps? Calculating real inflation-adjusted investment performance is not straightforward and ends up becoming necessarily subjective.1
There is also the problem of the long potential time-lag that can exist between large moves in currency exchange rates and inflation rates. Consider the recent example of the large devaluation of the UK pound sterling, which in the second half of 2008 lost about 25% of its value in trade-weighted terms. From the perspective of a sterling-based saver, their global purchasing power was in effect devalued by 25%. But has sterling inflation-linked debt risen in value by 25% relative to nominal sterling debt? Not at all. In fact, it has outperformed by only a modest amount. Why? Well, UK inflation has not picked up by that much. Currently, it is running at around 5% y/y.
The Bank of England currently anticipates that inflation will decline gradually over the coming year. This forecast may be wrong. Inflation could remain elevated and, as such, UK inflation-linked debt might have more room to outperform relative to nominal debt in the coming years. But it should be clear from this example that inflation-linked debt provides suboptimal protection against currency devaluation. This is due in large part to the fact that domestic wages normally comprise a substantial portion of an economy’s overall price level. Just because a currency devalues doesn’t necessarily imply that wages are going to come under material upward pressure. In the case of the UK, unemployment is high and the banking system is impaired. As such, it is likely that pressure on wages will remain muted.2
Of course, over a period of many years, if UK inflation remains elevated, inflation-linked debt might eventually outperform nominal debt cumulatively by a more substantial amount. But it would have been much better for UK savers from the outset to hold their wealth in other currencies which rose in value relative to sterling. It is much more attractive to realise a 25% gain up front, which can subsequently be reinvested as desired, rather than to wait for years for a cumulative outperformance of a potentially comparable magnitude. In any case there remains the uncertainty regarding whether the reference inflation index properly measures changes in the purchasing power of the domestic currency in the first place.
The same could be said about commodity price shocks affecting currencies generally. Were commodity prices to spike 25% higher overnight, it would be unreasonable to expect wages in most countries to follow quickly, in particular in a relatively weak economic environment with high unemployment and credit impairment. As such, inflation-linked debt would not outperform in anything like the same magnitude as commodity prices rose. For investors, it would be much better to hold the commodities outright, realise the 25% gain and reinvest thereafter, if so desired, in a broader range of assets. Which then brings us back to our original question: If neither depreciating currencies nor inflation-linked debt are reliable, stable stores of value, rendering them less useful (and potentially highly misleading) as benchmarks for measuring real wealth and investment performance, what are the alternatives?
For much of history, under some form of a gold or silver standard, global investors measured their wealth in terms of ounces of precious metal, at least indirectly, in that all major currencies represented explicit claims on various defined weights of these metals. However, since the US went off the gold standard and the Bretton Woods system of fixed exchange rates broke down in the early 1970s, this sort of indirect benchmarking against gold has no longer been possible. To benchmark against gold today requires that an investor do so explicitly.
For some, however, to measure their wealth and benchmark investment performance against gold might be a stretch. After all, gold no longer functions as a medium of exchange and is not recognised as legal tender. It is also an unproductive asset with no yield. And unlike the necessities of life such as food, clothing and shelter, you don’t need to "consume" gold.3 For some, gold is no better than casino chips that must be cashed in if they are to provide any economic benefit. But gold does have one important quality unlike casino chips and fiat currencies, which is that it is costly to produce. Indeed, the prohibitive cost of producing gold practically guarantees that the supply is likely to grow only slowly over time. No government can “print” gold. And while some can dig it out of the ground, the cost implies that it only makes sense to do so at an elevated price. And what might push gold to an elevated price? Looked at from the opposite direction, what might erode the value of fiat currencies? Well, we would focus on the primary concern we have discussed in previous editions of the Amphora Report, that numerous governments face unsustainable debt burdens and are going to default on and/or devalue these debts at some point in future.
However, notwithstanding its historic role and strictly limited supply, is gold really a sensible benchmark unit of account? Let’s take a step back. Why do we save and invest? Presumably to provide for future consumption in uncertain circumstances. Well, what exactly are we going to consume in future? None of us can know exactly but much of it is likely to include some combination of food, clothing and shelter.4 If there were some way to lock in a future price for those things today, would that be a reasonable way to think about real wealth preservation and investment benchmarking? We believe so. It may seem a bold intellectual leap to cease using the dollar, or other major currency, as the "denominator" of asset values, but by once you accept that that a currency is not stable, by corollary a new benchmark must be found.
WHEN IS A RISING STOCK MARKET ACTUALLY FALLING?
In domestic currency terms, the Zimbabwe stock market was a fantastic investment over the past decade. In some years it rose more than tenfold in value. In particularly good months it rose by several hundreds of percent. No major stock market performed in such spectacular fashion during this time. But this was not due to positive economic developments in Zimbabwe. On the contrary, it was reflective of the severe hyperinflation that was taking place. The Zimbabwe dollar, once a reasonably stable currency, began to depreciate rapidly in value in the early 2000s. Annual inflation climbed into the hundreds of percent in 2001, thousands in 2006 and hundreds of millions in 2008, prior to the complete collapse of the currency, which is no longer used as a medium of exchange.
Hyper-inflating currencies make poor benchmarks (Note the logarithmic scale!)
Source: Bloomberg, IMF
While we do not anticipate that the US or other major developed economies are going to experience hyperinflation, what if the dollar and other major currencies are no longer reliable stores of value, given policymakers’ proclivity to expand the money supply as necessary to provide support to weak economies and financial systems? Are major fiat currencies really the proper benchmark units of account? If not, what are the most appropriate alternatives? And how do such alternatives change our investment perspective?
***
If US policymakers continue to expand the money supply in a futile effort to prevent a natural and necessary deleveraging of the financial system and economy in general, this is a trend that is likely to continue, and quite possibly accelerate, in future. But given the frightening reality that in the euro-area, the UK, Japan and just about everywhere in the developed and much of the developing world there are large, growing and clearly unsustainable debt burdens, there are obvious limits on how much protection can be obtained through currency diversification alone. It is entirely possible that, at some point in the coming years, inflation rates are substantially higher just about everywhere as policymakers the world over succeed in devaluing their debts. There is quite possibly no fiat currency left in which to hide.5
According to available data on investment flows and, of course, the price itself, for an increasing number of investors, gold is the preferred safe haven. We would not presume to debate those who point out the impressive history of gold as a reliable store of value. But while gold most certainly belongs in a defensive investors’ portfolio, it is, after all, only one undiversified asset, and one which, at some point in future, is going to need to be exchanged for various goods and services. This calls into question whether gold, or any other single asset, is the most appropriate alternative unit of account for investors who have concluded that they should no longer benchmark their wealth against unstable, depreciating fiat currencies.
The Amphora Liquid Value Index (through 16 June 2010)
Source: Bloomberg
1 There is an active debate in economic academia regarding how best to calculate consumer and other forms of price inflation. Rather than enter into this debate we prefer to point out that governments periodically revise their inflation calculation methodologies, an explicit admission that previous methodologies were inaccurate. In the US, the calculation methodology was revised in the mid-1970s and again in the mid-1990s. At that rate, the next revision is due in 2015. Whether this might result, as all previous revisions, in a lower, rather than higher rate of consumer price inflation, is unclear.
An additional observation here is that inflation rates compound over time in the same way as interest payments on debt. So even in a country with a normally low rate of inflation, over long time periods the currency can lose most of its purchasing power. In the US, for example, depending on how inflation is measured, the dollar has lost from 70-95% of its purchasing power over the past 100 years. As such, revisions to the inflation calculation methodology that result in just a one- or two-percent change in the official inflation rate can, over time, compound into large differences in the implied overall change in the cost of living.
2 Most western economies have seen wage growth lag behind that of other prices in recent decades. Real wage growth in the US, for example, has been stagnant since the 1970s. Rising standards of living have not been due to rising wages but rather debt-financed consumption. Rising asset prices have, at times, provided the illusion of prosperity. While economic weakness and inflationary policy responses thereto are likely to contribute to weak real wage growth in the coming years, this is really only a continuation of a trend long-since established. Some commentators argue this is due primarily to increased competition from developing economies. We disagree. In our view, it is due primarily to persistently low and declining domestic savings rates over the past generation.
3 Gold does have some industrial uses but these consume only a tiny fraction of annual production. Some might consider the purchase of jewellery to be a form of gold consumption. Yet the production of jewellery does not reduce the amount of gold in existence, it merely changes it from bullion into some other, more artistic form. More accurate would be to say that, periodically, gold bullion goes out of circulation for use in jewellery production yet, from time to time, when gold (or silver) prices are high in relative terms to other assets, re-enters the bullion stock as it is melted back into coins (or bars) for use in exchange. Readers might have noticed that there are currently widespread advertising campaigns offering “cash for gold”, a sign, perhaps, that we are re-entering one of these periods, when the demand for bullion gold (and silver) is high enough to encourage households to part with family heirlooms in return for a fat wallet. There are also apocryphal (or not?) reports that copper pennies in the US, UK and other countries are gradually falling out of circulation as their melt value now substantially exceeds their monetary value.
4 Readers are probably well aware that in most developed economies in recent years, the costs of food, clothing and shelter have not risen by nearly as much as those for health care, private education and other professional services. Unfortunately, there are not direct, practical ways to lock in today the future costs of such services. However, costs of this sort tend to rise in line with wages over the long-term. If wage growth remains stagnant, as in developed economies it has been in real terms for decades, then it is likely that, in real terms at least, the costs of such services decline in future.
5 As this Amphora Report goes to print, China has made an announcement that they are going to allow a gradual appreciation of their currency, the yuan, which was re-pegged during the global financial crisis in 2008. In response, numerous Asian currencies rose versus the dollar today in the foreign exchange markets. While this may well continue, it is important to keep it in perspective. While Asian currencies might continue to rise versus the dollar, the dollar might also continue to devalue. Notwithstanding relative economic strength, we are not convinced that Asian currencies are going to provide stable stores of value against which to properly benchmark investment performance in the coming years.
