In my last article, I gave readers some themes that you can exploit for better returns in the US equity market. For today's article, I'm branching out into how you can apply some of these same themes to US Treasuries, for timing the S&P 500, and to hedge risk and add a source of return in the currency and commodity markets. These strategies tend to be used effectively by well-run hedge funds and ultra high net worth individuals, but are mostly ignored by retail investors.
1. Using risk parity and US Treasury supply/demand to make better return forecasts.
Research shows that shorter-term Treasuries usually have better risk-adjusted returns than long-term Treasuries, and that both make a great hedge for stocks. The trick to turning this into money in the bank is to use a concept called risk parity. A good working definition of risk parity is to take a low volatility asset and apply leverage to set it equal to the volatility of either the S&P 500 (for equity strategies) or the 30-year Treasury bond (for bond strategies). My historical data confirms that risk parity works as advertised. However, I think I've figured out some of the reasons behind why it works so it will work better in the future out of my test sample. Understanding regime shifts can help us make better forecasts. In my last article, I wrote that, for example, there’s a 12 percent chance of snow on any given day in Chicago, but knowing whether it's winter or summer makes a big difference if you're betting on whether it will snow. US Treasuries are one case like this. Shorter maturities tend to outperform longer maturities per unit of risk taken. However, like winter in Chicago meaning snow is more likely than in summer, it’s helpful to understand that these factors are likely due to the shape of the yield curve and the Fed Funds rate. Also, as always, feel free to follow the links below to validate my tests. I think it’s pretty interesting stuff!
2008-2019–A New Regime for Interest Rates?
Blue is the 30-year bond, red is the 10-year note leveraged to the duration of the 30-year, yellow is the 5-year note leveraged to the same duration, and green is the S&P 500.
Source: Portfolio Visualizer
Since 2008, leveraged 5-year Treasuries have beaten the S&P, which is great if you own both because Treasuries generally hedge equity risk in a recession. 2-year Treasuries have done as well as 5-year notes but aren’t on the graph above because the software only lets me test 3 portfolios at a time.
2015-2019–Interest Rates Since 2015- The Market Almost Perfectly Priced in Rising Rates
Source: Portfolio Visualizer
Since 2014, the returns have not been as strong but still have been respectably positive. The key to doing better is to take advantage of carry and trend factors, but also to pay attention to value, meaning not to buy as many Treasuries regardless of maturity if the 10-year real yield is negative. The real yield of the 10-year has a strong mean-reverting tendency, and your risk goes up in owning longer-term Treasuries when you're not earning at least the rate of inflation, so this helps keep your risk constant. Also note that shorter term Treasuries are a better hedge per unit of risk than long bonds, making them potentially more valuable as a hedge than their returns in a rising rate environment would imply. Conversely, if real yields are attractive the strategy can be allocated more risk.
1. Calculate the carry + rolldown return of 2-year, 5-year, 10-year, and 30-year Treasuries. The maturity with the highest 3-month expected excess return per unit of duration is the winning move. Then, to implement, use leverage to set their durations equal. You can also mix and match Treasury maturities to target different kinds of exposure. In an upward sloping yield curve environment, shorter maturities tend to win. Recalculate monthly.
2. If the 10-year real (nominal - inflation) yield is negative, reduce exposure. If the 10-year real yield is >100 bps, increase exposure. Since yield curves tend to move in parallel, this does a great job of improving forecast strength. Combined, this takes the Sharpe ratio of the US Treasury model to around 0.8-0.9.
3. A common criticism of risk parity is that it relies on the negative correlation between long-term Treasuries and stocks to deliver returns. If you can leverage short-term Treasuries, however, you can directly take advantage of Federal Reserve rate cuts.
4. To this point, short-term Treasuries with leverage tend to have less variance than long-term Treasuries. This is because the long end of the curve can move more than the short end of the curve in less time. The risk of rising rates is something you need to manage, but central banks are on your side, so rates tend to not rise too dramatically in developed countries.
5. Contrary to popular belief, I don't find that buying Treasuries when the yield curve is inverted is a great idea, at least not if it's causing you to have a negative carry. During times of inversion in the yield curve, your move is to either invest in cash or in long-term Treasuries, or both (assuming they've not inverted also).
6. The test assumes Treasuries can be financed at 15 basis points above the risk-free rate in the futures market. If you believe the SOFR will be higher, then this will affect your returns. The good news is that basic trading software can see the implied financing rate of the trades in advance so you won't be walking into a trap.
2. Taking advantage of the leverage effect to make better 1-3 month stock return forecasts
1. Future equity returns are notoriously hard to forecast, but 1-3 month future volatility is relatively easy to predict based on current volatility. This allows us to decide how much risk to take, the same way card counters decide how much to bet. If you assume that 1-3 month equity returns cannot be predicted but volatility can, this allows you to forecast the 1-3-month expected Sharpe ratio of the market - as it fluctuates between roughly 0.25 and 1.0 (the long-term average is 0.4-0.5). An easy way to gauge volatility and take advantage of the leverage effect is to use the 200-day moving average as a volatility forecaster (the leverage effect is the tendency of volatility to rise and investors to get forced margin calls when markets drop, and of companies to add demand for stocks with corporate buybacks and of speculators to aggressively buy when stocks rally and corporate profits improve, respectively).
2. Combined with the ability to use leverage through S&P 500 and Nasdaq futures, this allows us to translate higher risk-adjusted returns into higher portfolio returns, period.
Here's a quick test I ran on this (link to the full test below).
Source: Portfolio Visualizer
3. The Fed is more likely to cut rates when markets fall, so you see higher risk-adjusted returns for US Treasuries when the S&P is below its 200-day moving average. This makes Treasuries and stocks a natural pairing in this kind of model.
4. Combine these signals for Treasuries and stocks using a volatility forecasting model, and you can boost your portfolio Sharpe ratio to a surprisingly high figure.
3. Forecasting commodity and currency returns for hedging and standalone alpha
You will want to diversify your portfolio between US and international stocks. However, a common problem is that international stocks tend to have poor risk-adjusted returns compared to domestic stocks. You can solve some of this by using factor investing abroad just like you would at home (value, quality, low volatility, etc), but you can't solve the systemic risk due to currency fluctuations. The solution is to hedge the FX exposure of your international stock positions depending on the correlation between your investments and your home currency. Here's an illustration of 1. using factors in international stocks for US investors, and 2. hedging currency exposure.
Applying the low volatility anomaly and FX hedging to international stocks.
Source: Portfolio Visualizer
Currency forecasts aren't too hard to make, as they only need a few quantitative inputs and some qualitative common sense. Carry, trend, and inflation expectations are the main quantitative inputs, and central bank independence and basic political understanding are the main qualitative inputs.
Understanding how capital flows in times of market stress can help you avoid surprises. The idea is to diversify by country and understand your commodity exposure, and also to invest in liquid currencies. Your ideal global long/short currency exposure depends on your home country and base currency, so don't blindly apply this, but understand that you can enhance your portfolio returns by earning interest rate differentials in currencies and forecasting FX movements, and reduce your FX risk when you own international stocks by establishing the proper hedge ratio.
Commodity forecasts aren't that hard to make either once you understand a few key concepts. For crude oil, for example, you can use the term structure and implied volatility in options contracts to make pretty strong supply/demand forecasts (Sharpe ratio of <1.0). I wrote about this a couple of weeks ago. Several other commodities will give trading opportunities from time to time, also, based on the same indicators of carry, trend, and value.
I won't get into the full details of how to forecast commodity and currency returns because I don't want to bog everyone down, but you get the idea. Hopefully some of the concepts here you guys can use to improve your portfolios!
Did you enjoy this article? Consider following me for future research updates!
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.