The risk parity weighted asset volatility (the Worldvolatility) stands at 5,59%. This is clearly within the normal range since the early 1970s. The 8th low at around 4% since 1974 occured in 2014 and from this low reading a clear uptrend can be seen until today. This can still be seen as mean reversion and the Worldvolatility have many levels to pass on the upside before it can be said to be high. In other words: -We live in a “normal” world at the moment resembling the late 1990s.
Asset volatility is still increasing very slowly but surely and has been in uptrend since 2014. Trump winning the US election did not change the direction of volatility and the latest increase is mostly because of a treasury sell-off. Asset managers are still experiencing “the wind increasing slightly as time passes”.
The path towards creating a worldvolatility have now reached the stage of an assetportfoliovolatility. The portfolio assets are :US 10 year treasuries, NASDAQ and gold. The portfolio is risk parity weighted and weekly rebalanced to capture actual experienced voll by the biggest asset managers in the world.
The last little increase looks small but captures the uptick in volatility after Brexit. The next phase towards worldvolatility will be some method to capture changing monetary regimes, with the China SDR inclusion in late 2016 being an example of a change of this type.
The 50-50 Nasdaqcomposite-10y treasury continually dollar-rebalanced portfolio shows a range for the the historical volatility. 5% sets a floor in the post bretton-woods world. 20% annual volatility seems to be as bad as it can get at least in dollar terms. The equity market and the treasury market represent the most important claims in the financial world and in dollar terms they only deviate this much for the holder of the 50-50 portfolio. These claims can be said to be proxies for claims on the american public and private sectors respectively. For the holder of claims on both sectors the volatility moves in this range: 5-20%.
The historic check of Nasdaq and 10year-treasury volatility shows a range of normality for these measures. Treasury volatility seem to be “calm” around 5% and that sets somewhat of a floor. Nasdaq volatility seems to have around 10% as a floor at least in post-coldwar world from the nineties and forward. Upper level obviously harder to pinpoint. Both these series are slightly leptokurtic (fat tail distributed) with Treasuries being more leptokurtic over the 42 year period with only one big spike in volatility.
The series looks to be working in “alternating-fashion” in a way with big spikes in one being “safeguarded” by calm in the other. This being brought about by portfolio manager behaviour and central bank behaviour over different periods with different investing fashions in vogue, for example the Volcker-moves of the early eighties by Fed and risk parity investing today.
Is this a reasonable basis for a “worldvolatility”? After researching financial time series and various kinds of societal time series for 20 years my view is that it is.
S&P500 volatility changed to Nasdaqcomposite volatility as benchmark for dollar denominated equity volatility. Nasdaqvolatility is leading changes in S&P500 volatility on a number of occasions in the last 40 years that is the reason for the change. Just like S&P500 it has been moving upward in the last weeks. Treasury volatility moving close to historic lows and opposite to equity volatility in the last weeks. Uncertainty about Fed rate hike cycle might be somewhat less than a month ago and might contribute to lower treasury volatility. Both volatilities are always scaled to yearly volatilities for comparison.
After the Yuan devaluation and very difficult times for many emerging markets, the Volkswagen scandal hits and the equity volatility rises. VIX also seems to be finding a new higher normal around 20-25 after spikes >30. Treasury volatility calmer than equity in this environment and not reacting as much to the upside. One factor that keeps treasuries calmer is the fact that in Yuan terms the return is a little bit better after the devaluation.
“Changes in the volatilities of different asset classes are significantly positively correlated,” says Ray Dalio and it seems from the Picture above that there seems to be more than an ounce of truth in that. With a correlation of 0,96 anchoring the volatility of S&P500 and 10 year treasuries in the last 8 years, it seems quite uniformly dispersed across asset classes, at least for equity and treasuries. The absolute level of volatility is harder to explain, for example the peak in 2009. One of the main pillars of “Risk parity investing” is that the relative volatilities hold under different economic environments. (The relative volatility of S&P500 and US10Y is averaging 1,73 above, this implies an average structural duration for Equity of about 27,2 years. With the 10 year treasury having a duration of 9,1 years and volatility being proportional to the Square root of time. 9,1*1,73^2=27,2 years). Risk parity investing seems to make sense at least over this short period of time (2006-2014).Even if the absolute level of volatility is not needed in risk parity investing it is still very fascinating and might capture some societal dynamics unknown today.