Financial market volatility has increased since the advent of globalisation and is undermining the real economy. At the same time, it is generally accepted that the non-correlation or low correlation of the various markets allows for portfolio diversification, improving return for an identical risk or reducing risk for a given level of return. But all such matters require much closer attention.
The volatility of a price or return is commonly accepted to be representative of the level of inherent risk in a market. Accordingly, the higher the volatility of an asset or market, the higher the risk of investing in the said asset or market. Volatility is calculated by the square root of the price variance (the standard deviation), i.e. by measuring the deviations in the price vis-à-vis its average value over a given period. Accordingly, a security whose price has increased or decreased in a regular manner over a given period is said to show low volatility. Conversely, if it has decreased or increased in overall terms in an irregular manner, with significant and numerous rises and falls over the given time period, it is deemed to show high volatility. With good financial logic it is legitimate to expect higher returns from a high-volatility security in advance.
We shall not enter here into the debate as to whether a measurement that treats gains and losses in a market in an equal manner provides an appropriate indicator of the risk taken by the parties, given that the latter tend to be more worried about losses than they are satisfied with gains. We shall concentrate on the issue of the existence or otherwise of increasing and excessive volatility since financial markets were opened up and deregulated in the 70s and 80s, and on the diversity and non-correlation of the volatility required for effective portfolio diversification.
In a somewhat paradoxical manner it can be maintained that, in equity markets for example, although volatility itself is highly volatile:
- Its trend has not actually increased since the progressive introduction of financial globalisation, nor over a much longer time span, namely since the late 19th century;
- Following the opening-up and deregulation of financial markets, however, it has shown a very marked increase in comparison to economic fundamentals (GDP growth and inflation), which justifies talk of excessive financial volatility.
This paradox can be easily explained if we expand the concept of volatility. The result of the calculation will obviously not produce the same signal if we vary the chosen frequency (such as daily or quarterly variation) or the period over which the volatility is measured (over a week or decade, for example). If we select a daily frequency for measuring the variation in the Dow Jones index, for example, and a weekly time period to measure the volatility of this variation, the first assertion can be verified. This means that the volatility trend in the equity market has not grown since the 1970s or 1980s nor, for that matter, since the late 19th century. In other words, daily price variation, measured over one week, has not been any higher on average in recent decades than in previous times. However, measured in this way volatility was naturally much higher, notably between 1929 and 1932, in 1987 and in 2001-2002 than in other periods*.
Such a frequency and timeline does not allow any judgement to be made on the comparative volatility of financial markets and economic fundamentals, as growth and inflation rates are only measured at lower frequency and only vary significantly over longer time periods. Additionally, a quarterly frequency and a timeline of a decade make it possible to adopt a pertinent approach to this question. A study into the volatility of actual financial variables was carried out by P. Artus (“Flash” no. 41 from February 2004, CDC Ixis). The figures below have been taken from this study.
Let us first of all be clear that such volatility measured for growth (GDP) has been relatively stable since 1960, with lower levels seen in the latter period (1980-2003) in the United States, and since the decade 1980-1989 in France and Germany, for example. As for inflation, its volatility is a little higher and variable than that of growth, with a peak in the decade 1980-1989 due to the strong deflationary phase experienced internationally. Accordingly, in both the USA (1960-2003) and in France (1970-2003), growth volatility lies within the 1.3% to 2.7% range and, for inflation, between 0.6% and 4.3%, with very low historical levels for both in the latter period. Finally, short and long-term interest rates have not seen higher volatility than that of growth and inflation.
This cannot be said for exchange rate and equities volatility, which increased sharply over the period under analysis. The effective nominal exchange rate for the dollar (weighted against the main partner country currencies) has seen its volatility multiply by a factor of 5 between 1970-1979 and 1960-1969, clearly explained by the end of the Bretton Woods system.
Fundamentally, financial stability is a collective asset, one which is vital for the proper functioning of all decentralised market economies.
This volatility once again nearly doubled over the following decade, reaching a level of 16.8%, but falling back down to 8% over the period 1990-2003.
The real stock market index volatility in the USA ranges from 11% to 18.5% between the decades 1960-1969 and 1980-1989 and rises abruptly to 82.6% over the period 1990-2003. In France and Germany, a similar phenomenon can also be observed. We can therefore assert without fear of contradiction that over the latter period there has been excessive equity market volatility compared to growth (around 1.5%) and inflation (around 0.6%). If we compare equities volatility to that of dividends in order to select a fundamental more directly associated with equities, we once again see the excessive volatility in equity markets
This excess clearly poses the question of instant pricing in markets which, according to efficient market theory, should be valid indicators of the fundamental or equilibrium values of financial assets. When price volatility as recorded on the equity markets, for example, is much higher than that of growth, inflation or long-term interest rates, the spot prices set by supply and demand lose pertinence and can legitimately be considered at certain times to be the result of disruptive speculative bubbles.
Accordingly, very short-term financial volatility that has not risen tendentially over a very long period can co-exist with medium-term volatility, notably in equity markets, at a much higher level than that of the fundamentals for the latter period under study.
The fact remains that financial theory justifiably teaches us that good portfolio diversification allows such volatility to be managed when there is total or partial decorrelation between the various financial markets.
A good selection of diversified assets, for example, should entail lower risk – and therefore lower volatility – for the whole portfolio with equivalent return expectations. But the 1980s and subsequent decades have unfortunately demonstrated that such diversification only brings its benefits in calm waters, not at times of serious financial crisis, i.e. when it is least needed. During stormy times such as during major financial crashes, volatility in the various financial markets (private bond spreads, stock markets in different geographical regions, emerging economy currencies, etc.) show the marked tendency to correlate abruptly in an upwards direction. The benefit of diversification and the ability to manage relatively high financial volatility falls off sharply, or even completely disappears.
The question of (excessive) financial volatility cannot, therefore, solely be managed by appropriate micro-economic measures. It remains a question of global macro-economic regulation. Fundamentally, financial stability is a collective asset, one which is vital for the proper functioning of all decentralised market economies, and must be managed as such by national and international regulatory bodies.
* See “Revue d’Economie Financière” (no. 74, 2004), study by T. Chauveau, S. Friederich, J. Héricourt, E. Jurczenko, C. Lubochinsky, B. Maillet, C. Moussu, B. Négréaand H. Raymond-Feingold.
Measured over very short periods we have experienced financial market volatility that has not shown any upward trend between the late 20th century and today; however, measured over long periods, volatility has indeed grown and has far exceeded that of the fundamental variables supposed to determine the prices of the financial assets themselves. Here we look at the reasons and stress that such high financial volatility cannot simply be managed by portfolio diversification.