True investment risk
Several definitions of “market risk” are used in the investment world. Since the early 1950s, however, the dominant definition has equated market risk to the volatility of total returns. This, in turn, has led to the construction of “balanced portfolios” comprised of stocks – the more volatile, riskier asset - and bonds – the less volatile, safer asset - often in a “60/40” mix: stocks to generate returns and bonds to dampen the fluctuations of said returns. Such an approach has worked very well over the last 35 years, yet we believe that its success is more the result of an ideal market environment (mainly falling interest rates) than a smart way to construct a portfolio: in other words, most certainly it won’t work going forward. Which is why, at Harvest, we take a different approach to risk management: decomposing risk in its underlying drivers, with the ultimate goal of building robust portfolios.
Excess returns
Our starting point is simple. Any investor faces a basic, albeit crucial, decision: keeping her[1] liquid wealth in cash-like instruments (such as: US Treasury Bills, bank deposits, money market funds, etc.) or investing it in any of the “standard” asset classes (such as: stocks, bonds, commodities, real assets, etc….). Naturally, to give up the near-risklessness of cash-like instruments, in order to invest in riskier asset classes, the investor demands an excess return, over and above the “risk-free” return rate offered by cash.
And here lies a very important point, frequently neglected by investors: cash is an investment itself, hence competing with the “standard” asset classes for investors’ money. Consequently, any change in the expectations of the return offered by cash would change its relative attractiveness to non-cash assets, thus having an impact on their expected excess return, in turn inducing a change in their prices. And that is true risk.
Drivers of true risk
It is to better understand, and manage, the change in the expected excess return above cash that we decompose it in its three underlying drivers:
1. Change in expectations of future cash rates
2. Change in the risk appetite
3. Change in the economic environment
Change in expectations of future cash rates
Return on cash is, by and large, set by a country’s central bank, which adjusts short-term interest rates in order to stabilize the economy around its chosen target inflation rate (often set at 2%). Normally, this means that a higher growth in the economy and/or a higher risk of accelerating inflation are expected to induce central banks into raising rates, as they seek to prevent the economy from running too hot (there are other considerations influencing central banks’ rates policy, but they are of little relevance, here).
So, how do we defend ourselves against the risk represented by changing expectations in future cash rates? The key point about the rate of return on cash, is that it is a common component of the total return for all assets, hence any changes in the market’s expectations to future cash rates do impact all assets. Therefore, changes to cash return cannot be diversified by holding a portfolio comprised of different asset classes. The only protection against this driver of risk, then, comes from either “buying insurance” in the form of derivatives (put options, for example) or in tactically reducing exposure to the assets (or, more aggressively, establishing a negative exposure by shorting the assets).
Change in risk appetite
The next driver of risk is the market’s appetite for taking risk. Investors, explicitly or implicitly, try to match assets with future liabilities. Assets can be broadly categorized into “safe” or “risky”, depending on their capacity to match future liabilities, both in terms of solvency (will the money invested be returned?) and timing (will the money be available at the very time it is needed?). As an example, if we needed the money for an expense due in three years’ time, both cash and a 3-year government bond meet both requirements of solvency and timing. A 10-year government bond, on the other hand, would satisfy the solvency criterion, but would fail the timing one, as it might well be trading below par when we needed the money in three years’ time. Further along the spectrum, planning to fund the expense in three years’ time by investing in equities would be a failure on both criteria.
Given the above, cash and government bonds (whose duration matches future spending needs) can be considered as “safe assets”, while equity and credit have to be considered “risky assets”. The historical average mix of safe and risky assets is then used as a proxy for the normal risk appetite of investors in a given country. Whenever the allocation to risky assets is higher than normal, risk appetite is deemed to be high and vice versa.
As illustrated by the chart above, the risk appetite fluctuates over time between what is commonly called “risk on” and “risk off”. Importantly, when risk is “on”, it can be risky to hold risk assets as it does not take much for investors to turn a little bit more conservative and cause a sell-off in risk assets. Investing in risk assets when risk appetite is strong means betting that either it will remain strong over the full investment horizon or, in case it dropped, it would cycle back up before the end of the intended holding period.
What is, then, the possible defense against the risk represented by investors’ changing risk appetite? In this respect, there are strong similarities with the first driver of risk, the changing cash rates: risk appetite, too, is ubiquitous for all “risk” assets, hence it cannot be diversified away by holding a portfolio of different classes of “risk” assets. The solution, unsurprisingly, is equally familiar: the only protection against this driver of risk comes from either “buying insurance” in the form of derivatives (put options, for example) or in tactically reducing exposure to “risk” assets while increasing exposure to “safe” assets (or, more aggressively, shorting “risk” assets while going long “safe” assets).
Change in the economic environment
The change in the economic environment affects assets’ returns via two main drivers: inflation expectations and economic growth. Luckily for investors – and a significant departure from the other two drivers of risk - inflation and economic growth impact different asset classes differently, making it possible to combine the various asset classes in a way that results in structurally diversified portfolios.
Given the critical role that structural diversification plays within our investment approach, it is worth looking at inflation and growth’s differential impact in a little more detail.
Inflation is a critical factor for our investments since, if we aim at maximizing the purchasing power of our wealth, we want to be compensated in real terms for the risk we’re taking.
Economic growth also influences investors’ decisions. When we expect growth to be very weak we demand a lower price to buy equities, as some protection against their likely future declines; yet we are willing to accept lower yields from the bonds, as they offer some safety for an uncertain future. Vice versa, when we anticipate strong growth, we are willing to pay a higher price for equities or demand a higher yield on the bonds, as some protection on the likely fall in bonds’ prices.
So, to get to the core of the matter, how do inflation and growth impact the main asset classes?[2]
Let’s start by rising inflation. A rising inflation, of such a nature as to make likely an increase in rates by the central bank, would definitely be negative for bonds’ prices. This is because investors would require a higher yield to hold bonds, in order to: 1) make bonds competitive again with the now-higher cash returns and 2) compensate the bond holder for the increased erosion (due to higher inflation) in the real value of the capital she will receive upon the liquidation of the bond. Yet, given that in a fixed-income asset the interest rate is, by definition, fixed, the only way to offer bond holders a higher yield is by dropping the price of the asset.
Equities, from a qualitative point of view, are affected by rising inflation pretty much the same way as bonds, in the sense that they, too, need to offer a higher yield to persuade investors to hold them. The difference with bonds, however, is that unlike bonds, equities can “recover” part of the damage inflicted by the rising inflation by passing on to their customers at least part of their increased costs. The net effect is that meaningful inflation increases have a milder negative impact on equities.
Commodities are supposed to do well in, or at least not to be negatively affected by, inflationary environments, mainly because “they are the inflation”, being themselves relevant constituents of inflation indices.
Moving on to the impact of economic growth, now, we can safely say that a robust economic growth is definitely positive for equities as an asset class, for pretty obvious reasons. Likewise, for commodities: a strong economy is expected to result in a strong demand for commodities.
Conversely, a robust growth is negative for bonds, pretty much for the same reasons mentioned above when we looked at rising inflation: a strong economy normally increases the likelihood of rising inflation, with the ensuing dynamics mentioned above.
To summarise the likely impact of inflation and growth over the main asset classes, the table below could serve as a good illustration:
The last step, in order to build portfolios that are structurally diversified - hence robust to the unpredictable market changes - is therefore pretty simple (although by no mean easy): to combine the various asset classes, according to their sensitivity to inflation and growth, as identified above. As examples, government bonds offer a good hedge for equities with growth, but not with inflation. Commodities offer an inflation hedge for equities but no hedge against falling growth expectations. A portfolio comprised of commodities and government bonds is balanced across different economic climates.
Conclusion
We hope that both the rationale and the benefits of our approach to risk management are now clearer: it is only by decomposing each asset class into its drivers of return and risk that we can construct robust portfolios, capable of withstanding the full range of unpredictable economic climates, particularly the most adverse ones.
[1] In our writings, to express neutrality towards either biological sex (and to avoid risible spelling contortions), we alternate the masculine and the feminine, selected on a random basis. For this paper, the choice has fallen on the feminine, which will then be used throughout the text.
[2] Please note, a detailed answer to this question is beyond the point of this document hence, here, we will describe only the main relationships between the variables in question.