Credit spreads refer to the difference between the yield on one particular instrument (or basket of securities) versus another. Spread trades can be among the most lucrative because you can calculate your risk/reward with some degree of precision. If the spreads between different types of credit securities contract more than what can be sustainable in the long-run, you can identify high reward-to-risk opportunities. How do you know what’s sustainable or not? It’s predicated on the loss-given-default calculations. All financial assets of various risk tiers must retain appropriate risk premiums relative to each other over time. Here are some examples: _ 1) BBB-UST Spread BBB-rated credit is in the lower investment grade (IG) range. UST Treasuries (USTs) are considered among the safest financial securities you can invest in globally given their backing by the US federal government. A certain amount of BBB credit is expected to default. Accordingly, there should always be some spread between the two when comparing comparable durations. There are also constraints to how tight this spread can go. The BBB-UST spread is currently 154bps. That is, lower-grade IG credit gives you an extra 154bps of annualized return over Treasuries. Let’s say things go well for risk assets and this spread tightens by 50bps. This would bring it down to 104bps, to a point even tighter than the 2006-07 period. In recessions, expect this spread to blow out to 300+bps. In December 2008, the nadir of the credit markets during the previous recession, this spread expanded to 800bps. Accordingly, your reward-to-risk is close to 3/1 in this situation if you bet on that spread based on (300 - 154) / (154 - 104), going by a conservatively estimated upside spread minus the current spread divided by the difference of the current spread and estimated downside. Taking this spread at below ~125bps is ideal. This enhances your reward-to-risk non-linearly. This spread can only tighten so far based on realistic expected loss calculations using standard BBB credit metrics. USTs and IG are less volatile than high-yield (HY), controlling for duration. This means you can responsibly leverage this trade more than betting on a spread widening between HY and USTs. How do you bet on this? Long USTs, short a basket of BBB corporate credit. I generally prefer to do long bonds while shorting futures for the corporates. Also note that if a futures market is backwardated and you’re short you can expect to pay the roll yield. Other possibilities include futures options, equity options (i.e., on the ETFs), shorting the underlying bonds, or shorting ETFs. (The main problem with ETFs is usually the relatively high margin requirements. Also, when shorting ETFs or bonds you pay the carry.) Long-dated options tend to be best when an idea is not time-sensitive. Sometimes structuring trades as options can be helpful to limit downside and it’s generally helpful to be long gamma. But other factors that price into options, like vega and theta, might not be part of the idea and you might not want those. IG corporate credit futures can be located through IBXXIBIG on CFE. _ 2) HY-UST Spread During this cycle, the B-rated credit (part of high-yield – i.e., HY) spread bottomed at 328bps on Oct-2-18. In June 2007, this compressed down to 241bps. It’s currently 357bps. HY has a much larger loss given default than IG. In most recession, the HY-UST spread will expand into the 800-1,200bps range. In a bad recession, it can run 1,500+bps. Per conservative estimations, let’s say your risk is a tightening to 225bps and an upside of 800bps. Reward/risk is then calculated as: (800 - 357) / (357 - 225) = 3.4x How do you bet on this? Same concept, long USTs (or a basket of the highest-grade IG) and short HY. If you already have a lot of exposure to USTs or IG, you might choose to short only HY. HY futures can be found through IBXXIBHY on CFE. ___ The Idea Behind Shorting Corporate Spreads Spread trades are a staple because it can be easier to calculate your reward-to-risk relative to directional trades. It is possible to calculate how much a spread can contract. You can’t just look at a previous chart, like those displayed above, and use that as your theoretical bottom. That’s a lazy way to do it and not particularly logical because the future can be different from the past. You have to do the calculations and not simply take a guess. But as aforementioned, you know certain things like if the forward yield on the 2-year US Treasury is within 150bps of the forward yield on stocks, you’re likely not making a very good reward-to-risk trade being net long equities. (For the record, this spread is currently 320-330bps.) As a past example, during the Oct 2002 to Dec 2007 business cycle, CDS on 10-yr European sovereign debt was trading at 5bps. When you purchase a CDS your loss is simply the annual interest rate over the life of the bonds – so 5bps per year over 10 years. When the economy went into recession, this spread increased to more than 75bps. In normal recessions, you are likely to see these spreads increase to 40-60bps. As a result, you know your reward-to-risk on this particular trade when the spreads are this tight is probably 8-to-1 or better. It’s also a risk-off trade. Portfolios as a whole are too concentrated in risk-on bets because those are the ones with the highest expected yield. So you usually see a lot of equities. Having risk-off trades, when you can identify them, acts as a form of insurance with limited downside and a low premium. Plus there’s the benefit of asymmetric upside where your reward is several multiples of your risk when material and relevant changes occur to the economy and markets. Risks If there is no credit contraction in the economy (still at least a year away in the US), you can sit in a neutral or losing trade for a long time. It requires patience. And returns are also time-sensitive. Even if you earn 35% but it took four years to play out, that return may be too low relative to what you expect. Interest rate risk comprises nearly the entirety of the movements in USTs and the safest forms of credit. Sometimes this factor dominates credit risk and you lose money. This is typical in late-cycle economies when rates are rising but credit risk isn’t increasing in conjunction. Catalyst Your catalyst is a compression in private sector credit creation when defaults spike, creditors pull back lending activity to limit losses, and debt servicing costs exceed the incomes, assets, and new credit available to pay the debt. Important Points 1) The reward/risk of these trades improve non-linearly due to the nature of this calculation. The numerator increases, while the denominator decreases. It follows in this format: (Upside Spread – Current Spread) / (Current Spread – Downside Spread) It’s also a risk-off trade. Spread trades, in general, are uncorrelated with the broader market. Slowing global growth and the pullback of liquidity by the Fed makes risk assets riskier. The eventuality of higher financial markets volatility and higher term premia support the idea of being short credit spreads when your reward/risk is high enough. The big widening in spreads won’t occur until there’s a debt problem in the economy. 2) There’s a notion you see in trading that because you have a take-profit level further away from the current price level relative to some stop-loss that’s being employing that this represents a high reward/risk trade. However, this isn’t true by itself. If there isn’t a catalyst to get price to the targeted level, that’s not actually a high reward/risk trade at all. For example, if someone goes long Google stock (GOOG) at $1,050, sets a stop-loss at $1,000 and a take-profit at $1,250, they might view that as a “4-to-1 reward/risk” trade. For this to be a believable and informed claim, they would need to identify what the catalyst is (ideally with some math behind it) beyond the basic beta return associated with being long equities. (Note: I don’t use stop-losses. If risk premia are expanding it makes sense to add, not subtract.) Reward/Risk Curve Based on the above formula, each unit of contraction in the spread has a positive exponential effect on your reward relative to your risk. This goes for all trades, not just what’s covered in this particular article. Understanding risk and reward is fundamental to trading in the same way it’s fundamental to games like poker. You need to be able to clearly identify within reasonable certainty how bad something can get and compare that to how good something can be. And then using that information to bet when the odds are in your favor and don’t when the odds are against you, neither favorable nor unfavorable, or unknowable. It is also why it can make sense to add to a trade in higher amounts if a position goes against you, so long as what occurred doesn’t change your mind about whatever you’re trading. It’s because you improve your reward/risk and this will be reflected by improving your cost basis. Map out your game plan, set your prices, start small, and, if necessary, you can work your way up. General Conclusions When it comes to any financial asset, its price is simply the present value of a stream of future cash flows discounted back to the present. Over the long-run assets’ cash flows are fairly stable and predictable when looking at the level of broad indices. But their present values (their prices) will fluctuate. Paying a higher price for a trade necessarily means lower future returns. If the price of an asset goes up, returns in the future are being pulled into the present. This means there exists a mechanical inverse relationship between realized returns and expected future returns. This is going to involve some calculations. It is also important to fight the natural tendency to buy into stories or go off what one reads or hears. Many people invest in stories because it’s easier to grasp onto them. But that’s too imprecise. It’s important to quantify matters and try to understand why assets are cheap or expensive by using measures of what cheap or expensive actually means. Some things you may not know and will need to estimate conservatively. Then fight the tendency to trade stuff you have only marginal conviction in. Or fear that if they don’t get in, then it might go in their favor and it will be painful to miss out. (It is worse to lose money than miss out on money you may have gained if you had made a different decision.) People tend to make decisions if they believe their odds of being right are greater than 50/50. But your odds can be raised – i.e., their reward-to-risk – to an even higher number by either changing their entry price or obtaining more information. Only get into the trades you really like. There are a lot of markets available. And you also don’t need to fill up your portfolio with a bunch of different low-conviction stuff. If you can find 10 genuinely uncorrelated and equally good bets you will improve your reward/risk by more than 3/1 over simply being invested in one asset. In other words, you won’t materially hurt your returns while eliminating two-thirds of the risk. If you can find 20 of these, then your reward/risk will be improved by a factor of between 4/1 to 5/1. But there are diminishing returns the higher you go. Graphically the relationship appears as follows: The higher to the left you can be in the graph in the preceding section and the more to the right you can be in the graph above, the better off you’ll be.