Click here for Analysis Report Archives

Download .pdf file

January 2007

The sharp decline in global volatility since around 2002 has come in for much analysis and comment.

A decline in volatility is, in some senses, the quantification of a decrease in risk aversion or a rise in risk appetite. In this report, we have used the VXO index (which is a forward looking volatility of the S&P100) as a surrogate for these.

The VXO fell from around 50% in 2002 to 22% by the middle of year 2003 and has slipped slowly from that level to around 11% today, an over 20-year low. What is more remarkable is that it has stayed at these depressed levels for over 3 years, which is certainly a record. The phenomenon of low volatility is not specific to equities but has taken place across all widely traded financial assets, including G-7 currencies and bonds.

There have been many reasons put forth to explain this unusual phenomenon of low volatilities

• At the macro-economic level, the recent fall in asset price variability may reflect the ongoing phase of sustained expansion and low inflation experienced by the world economy. Globalization, in other words, may provide some of the reason.

• Another structural parameter thrown up to explain lower volatility is the so-called "great moderation"; which is the fact that, since the mid 1980s, output growth has become less volatile, both for the US and the G7 economies. The period also saw adoption of a variety of macro policy frameworks that supported macro-economic stability. These include inflation targeting adopted by some central banks, the adoption of fiscal frameworks such as "stability and growth pact" in the euro area and so on. Lower volatility in inflation has allowed policy volatility to decline. One important element has been the trend towards "gradualism" (policy synchronization- decline in volatility of interest rates across G7 economies) in monetary policy action. However, given that the great moderation began long before the sharp decline in asset price volatility suggests that this is an unlikely candidate for recent decline in volatility. By the mid-1980s, the decline in volatility of macro fundamentals had already run its course.

• The increased credibility of central bankers - the Greenspan era, which spanned 15 years ending in 2005 - could be part of the story. Volatility is basically a measure of risk, or uncertainty, and (so the argument goes) as markets have become more able to forecast the Fed's behavior - and, as a result, future interest rate movements - the risk premium they would demand for holding on to assets would fall.
  
  While this does make sense, the truth is that in recent months, there has been a dramatic increase in uncertainty about not only U.S. interest rates - well-considered forecasters are split on whether rates will rise or fall in 2007 - but about many fundamental macroeconomic parameters that would determine growth rates in the G-7 countries.
  
  Further, over the past several years, the world has been faced with a wide range of remarkable imponderables - from the ill-fated U.S. war against terror and its associated sharp rise in oil prices, to other geopolitical trauma (North Korea), to the rise of China and its now $ 1 trillion hoard of foreign currency reserves (held largely in U.S. assets), to dramatic and high-impact changes in global weather patterns (the Asian tsunami, hurricane Katrina, and so on), the swing to the Left in many countries (most recently and remarkably, Hugo Chavez' move to nationalize assets in Venezuela, the fifth largest oil producer in the world) - any of which would have been enough to trigger a retreat from risk in the past.

• There are many who argue - correctly in our view - that recent developments in the financial markets have contributed to the phenomenon of low volatility. Improvement in market liquidity, rapid growth of the market for risk transfer instruments and growth in the fraction of assets held by well-informed agents managing diversified portfolios, are all forces that would increase the ability of markets to absorb shocks - note how the collapse of the hedge fund Amaranth with over $ 6 billion in losses passed through markets without causing any sort of trauma (in comparison with the $ 1.6 bn collapse of LTCM in 1998, which resulted in a huge spike in market volatility and risk aversion).

• Finally, there are some arguments that believe the recent fall in volatility may have been a result of firm-specific components related to the improvement in the balance sheet conditions of listed companies: a decline in corporate leverage and an improvement in actual and expected profitability. Surveys suggest that the degree of uncertainty surrounding firms' profitability also decreased.

While all of these arguments do carry some weight, neither any one of them nor the combination provides convincing evidence that this decline in volatility is structural rather than cyclical - much of the academic research remains divided.

In this paper we argue the middle path - that some of the decline is, indeed, structural, considering the several points raised above, but some part is certainly cyclical. We also look at what forces could drive the cyclical volatility upwards and try and forecast what the impact would be on financial markets, particularly the dollar, and hence, the rupee.

Strong evidence of cyclical volatility decline

Perhaps the most striking evidence we have found of the cyclical nature of the recent decline in volatility is the strong correlation between the Fed funds rate and the volatility of the S&P100. The chart shows that the correlation between the 1Year moving average of the Fed funds effective rate lagged two years and the VXO (CBOE S&P 100 Volatility Index), which started to rise around 1989, strengthened through the Greenspan years, and today appears to be very close to 100%. If this correlation continues to hold, it predicts that S&P volatility, following Fed funds, could rise sharply over the next two years, reaching levels last seen in 1996-2000, around the period of the Asian crisis, the Russian default and LTCM.

There are a few technical reasons that could support this forecast. First of all, it is intuitive that the volatility cycle lags the interest rate cycle. For instance, the Fed began raising rates to prevent overheating of the economy and a rise in inflation; its efforts have been successful, but rates remain high even as the economy has slowed down, which implies a reduction in risk appetite. Further, since there is currently increased uncertainty as to whether or when the Fed will cut rates - market is discounting no change in rates all the way out to March - it is possible that the shift away from risky assets could increase, pushing volatility higher.
The second technical factor is the absolute low level of short-term interest rates. We ran a simple regression of 1Y Fed funds rates volatility versus the absolute level of the Fed funds rate and found that since 1965, the correlation is close to 70%. This also suggests that if/when interest rates rise - as they have over the past year or so - there could be an increase in volatility with a lag.

Impact on US equities

Significantly, this increase in financial market volatility doesn't necessarily mean that U.S. equity markets will tank. Bull and bear cycles in the US equity markets are generally independent of market volatility, although sharp spikes in volatility are associated with market meltdowns (the three Black Mondays - October 19, 1987, October 27, 1997 and August 31, 1998 as also the meltdown on the New York Stock Exchange when it reopened after a week-long forced closure following 9/11. Since we are currently only looking at a steady rise in volatility - not a spike - we do not believe there is evidence to conclude which way U.S equities will move.

However, the impact on Indian equities (and those of other emerging markets) and the rupee may not be so benign.

Impact on the dollar

The correlation between the 6M moving average of the Fed's nominal trade weighted dollar index and financial market volatility has been very high over the last 10 years. An increase in financial market volatility has usually been associated with a sharp rise in the dollar - probably the safe haven effect.

This was particularly loud in the period of 1996 to 2002, despite the growing cacophony about the U.S. twin deficits. However, it is worth noting that there can be sharp contra-cyclical movements. But, in our view, these would be the result of short-term forces; the overall trend would be that if/when financial market volatility moves upwards, the dollar will rally sharply.

Impact on the rupee

A sharply stronger dollar would normally suggest a sharply weaker rupee, both because of the increased linkages between the Indian economy and the world, and also because it is likely - as we argue later - that the trigger for increased market volatility could well be another round of emerging market risk aversion.

We looked at the historical volatility for the Indian rupee since 1990. Current volatilities have been low, as compared to the levels last seen in 1991-92, when reforms were introduced. They are also low when compared to the levels seen at the time of the Asian crisis in 1997-98. Nonetheless, current rupee volatility (around 5%) is close to its long-term average, which is quite different from the current volatility levels for other currencies like EUR/USD and USD/JPY. While the absolute levels of these are, unsurprisingly, higher, they are quite a bit below their historic averages.

We also note that rupee volatility has increased recently, and the increase in volatility since 2003 is strongly linked to the sharp increase in FII inflows into the Indian capital market.
Of course, the higher volatility of the rupee due to FII inflows has also resulted in rupee appreciation - our model shows clearly that FII inflows are one of the long-term variables forecasting long-run movements of the currency (Forecasting the Rupee, Nov, 2006). Thus, FII flows directly influence rupee volatility and, quite obviously, will affect the level of the currency depending on their direction.

Impact on Indian equities

Coming back to global volatility, we have also studied the 2 year rolling correlation between VXO and 1M historical Sensex volatility, which has been running high for the last 1.5 years. Though the relationship is unstable and breaks up easily, it is interesting to find that the correlation has been positive for more than 60 percent of the time since 1988.

While this does not provide conclusive evidence, it does suggest that there is a strong likelihood that if/when S&P volatility moves higher; the Sensex volatility will rise as well.

In sum

A change in the cycle of global market volatility, which seems likely some time in 2007, would lead to a stronger dollar globally and a weaker rupee. Given the strong correlation between the rupee and the Sensex - the 2 year rolling correlation between USD/INR and 6M historical Sensex volatility has been close to 80 percent since late 2005, and that one of the impacts of the change in the volatility cycle would be a rise in risk aversion, we would also expect the Sensex to decline in this event.

But what could change the volatility cycle?

As we all know, markets sometimes appear to react to nothing at all. However, there is always one - or most, frequently, several - factor playing behind the scenes which acts as the proverbial straw that breaks the camel's back. The list of straws below, while fairly exhaustive, is presented in what we believe is the order of greatest likelihood:

• Emerging market risk aversion: Recently we have seen volatility in emerging market currencies and stocks. Asian central bankers are getting increasingly uncomfortable with the steady to strong appreciation of their currencies. Events such as Thailand's ill-advised foreign investment laws, and South Korea's more creative efforts to favor foreign outflows, together with Venezuela's proposed nationalization of utilities and the decline in commodity prices could shake investor confidence in the emerging economies' outlook. This could be a key potential source of increasing volatility in the FX market, since emerging market and private equity investments remain the only high volume risk-seeking bets in town (now that both the commodity and hedge fund plays appear to have run their course).

• Private equity problems: In recent years, there has been a huge increase in assets held by private equity firms [see box]. It is conceivable that part of the decline in volatility (risk aversion) could be related to the fact that market measures of volatility do not take into account the risk on assets that are held outside the market. With the scale of these holdings rising, any problems in a large private equity portfolio could tip the balance. Granted that most private equity investments are backed by strong management capabilities, but then let's not forget that LTCM was backed by the best minds in the financial risk management business.

• Carry trade unwinding: Stronger-than-expected economic growth in Japan could lead to a disorderly unwinding of the yen carry trade. When the BoJ raised rates the last time in July, for the first time in six years, the market was ready, and was virtually unanimous in predicting both the timing and the scale correctly. Since then, there has been a substantial increase in the size of positions and a great deal of uncertainty as to whether and when the BoJ will raise rates again, and, importantly, how much rates will rise in total. At its January meeting, three of the six board members voted against the decision not to raise rates, as a result of which the future markets are now discounting a 69% chance of a rate hike next month compared with 42% before the meeting. However, it is not just the next hike that is important - the real issue is whether Japanese interest rates will rise far enough (or, enough people in the market believe it will rise far enough) to trigger a sharp unwinding of carry positions, which could trigger a turn in the fundamental volatility cycle.

• Rising rates in Europe: There is a growing sense that central banks have less and less influence over financial markets, given the huge and rapid growth in volumes of structured products, which are neither readily measurable nor responsive to changes in monetary policy. As a result, central bank credibility, which is one of the key pillars upon which the decline in volatility has been based, could be weakened. The surprise hike in interest rates by the Bank of England in early January could be a harbinger. More such surprises could trigger nervousness and increased volatility.

• The no-landing scenario for the US economy: With Mr. Bernanke riding the waves of public acceptance surprisingly well, the US remains in a no-landing scenario - fairly stable economic growth with a Fed poised to act against inflation. While it is likely that this will remain the scenario through much, if not all, of 2007 - viz., rates steady but biased upwards - the changing economic landscape could increase market confusion, which could lead to a sharp sell off in the dollar and a spike in volatility, as we saw last Thanksgiving.

• Recession: The probability of a hard landing in the US, though low, is still significant. Recent data releases have portrayed a confusing picture of the US economy - for instance, it is still unclear whether the US housing sector has bottomed, while the US labor market remains still surprisingly strong. There is also evidence of the linkages between various economic variables breaking down. For example, in Germany there is a large disconnecting between the IFO and ZEW analysis on the expectations of economic prospects. In the US itself, there are large differences in the opinions about fourth quarter US GDP growth and a great divergence between the forecasts made by several investment banks on the outlook of US interest rates. On the forecaster front, there is a huge overhang of doomsayers combined with a steady diet of optimistic market forecasters. In other words, uncertainty is high, which, if exacerbated by a surprising dip in growth, could lead to broad-based reductions in risk appetite. Again, there have been record amounts of low-quality debt issued over the past few boom years (LBOs, private equity funding, etc.) and a recession, or even near-recessionary conditions, could likely tip some part of this huge iceberg into default, which could trigger a run to safe havens.

Could it be "different" this time?

History has shown that in markets the "this time it's different" hypothesis seldom works. Technology-driven structural change, increases in knowledge and risk-bearing capability of markets all notwithstanding, we believe that the historic low in risk aversion in the face of continuing mounting global uncertainty will not remain in place forever.

While it is impossible to predict exactly when things will turn and/or what will be the trigger, we believe there is a good chance the turnaround will begin some time in 2007. In terms of action, this suggests that the medium term bull trend of the rupee may see a sudden wobble, which could trigger a sharp correction. Out-of-the-money call options are cheap today and, we believe, worth buying.