Most people are familiar with the so-called Equity Risk Premium (ERP): the idea that equities generally return more than the “risk-free” rate because investors must be compensated for the added fluctuation risk and bankruptcy risk that they are typically accompanied by. Many extend this logic to bonds in general, claiming that equities outperform not just government bonds but corporate bonds, too.
This apparent phenomenon is treated as fact. Everyone knows that equities return more than bonds, that’s why you should just invest in an equity index and get back to work (apparently). As always, retail investment/personal finance gurus are heavy on the recommendations but not on the explanations.
Is this really true? What about the ERP?
I couldn’t find satisfactory answers to these questions.
Apples and oranges
Equities are fundamentally different from bonds.
Bonds are debt, issued by the company (or government) to finance company operations. Bondholders expect to be paid a fixed amount and are reaaaally not happy if they lose their money. Their exposure is asymmetric, in a bad way. It’s generally all downside and no large upside.
Equities are a different breed. When you purchase a share you are literally buying a piece of the company. Equity holders receive some of the company profit and capital gains if the company increases in price. Companies can go up a lot in price very quickly.
So, why would anyone hold bonds over equities? Firstly, because holders of corporate debt (bonds) get preferential treatment when it comes to recovering investment if a company goes bankrupt. Additionally, bondholders receive a fixed payment (as well as their original investment at maturity). If the company is struggling, it doesn’t matter. Bondholders still expect to get paid the same amount.
Risk and return
In financial markets, there seems to be some kind of relationship between risk and return, it’s just not known what exactly it is. But I think we can say with some kind of confidence that the relationship is in some way inverse: increasing return is associated in some kind of way with increased risk.
Equities have both more fluctuation and bankruptcy risk. This needs to be compensated for with higher total return (or the possibility for higher returns) relative to bonds. This can be achieved by paying a higher dividend (relative to bond yields) and/or price increase potential (capital gains). Investors must simultaneously weigh the risks and rewards of each proposition.
Prices constantly adjust to in an attempt to find the right balance. Imagine if bond yields were higher than dividend yields for some period of time for some company. Some shareholders would re-asses bonds as now being a better risk-reward proposition than equities. These investors would sell shares and buy bonds to capture this higher yield. This would increase the price of bonds and decrease their yield. At the same time, shares might drop in price (and increase in yield). Now shares might be more attractive and some bondholders may start switch to shares…
The market continuously self-corrects through this relenting mechanism of price discovery.
Does this mechanism mean that the total return of both the corporate bonds and the equity of every company should be essentially the same? Not necessarily. Firstly, it’s very difficult to price things such as risk (both bankruptcy and fluctuation), future economic variables (interest rates), and the future prosperity of the company (affecting bond payments, dividends, and capital gains). Even with the wisdom of crowds/semi-efficient market hypothesis, it’s still difficult to determine the intrinsic value of both equities and bonds. This is especially true when considering assets in perpetuity (the long-term is underpriced).
Secondly, there may be a mismatch in investment horizon between bondholders and equity holders. It may be the case that bonds are a better short-term investment (and have higher total return when risk is taken into account) and equities are a better long-term investment.
Finally, investors want certainty. There is something about not knowing that’s uncomfortable. It’s much better psychologically to receive a constant, fixed, positive drip than an unknown future payoff. It’s addicting. Certainty may also be required for cash flow reasons. Bonds provide this certainty.
Does this mean that equities are a better investment? In the short-term, it’s impossible to say. In the long-term, there does seem to be reasons as to why this might be the case: fluctuation risk becomes less of a factor; certainty becomes less valuable; and the long-term is generally under-priced.
We can use history (with a pinch of salt) to check if this might be true.
% CAGR for S&P 500 and Baa-rated Corporate Bonds indices. Remember, this is only looking at one time segment in one market. Also, we are seeing an aggregate, not individual companies. Finally, taking too much stock in this data is associated with all the problems of induction as well as statistical inference. Source: NYU Stern School of Business.
The annual returns are a mess, giving no indication as to which might be better. The 10-year rolling chart is more clear but there is still no clear winner. At longer time horizons, however, equity does seem to return more. But this gap seems to have narrowed recently. Maybe corporate bonds have become less popular (hence higher-yielding)?
Is it possible that this relationship reverses? Will bonds enjoy a premium at some point in the future? This seems unlikely. What has to happen for this to be the case? Equities would have to be a very unattractive investment, either not growing (no capital gains) and/or not paying any dividends (no yield). But if this happens, surely everyone would buy bonds and thus their prices would increase, lowering their yield?
This is the problem with these types of arguments. Due to the self-correcting characteristic of the market, it’s difficult to imagine a sustained period of out-performance from either shares or bonds. But this is exactly what we seem to have for equities. This is somewhat of a puzzle. The reported reasons for this out-performance are weak and it doesn’t feel like they entirely explain the phenomenon. At the same time, it’s hard to imagine this relationship reversing. There is no apparent reason for that to happen.
But I’m not ruling it out. Things happen all the time that are unexplainable. In theory, it’s unlikely but theory isn’t the same thing as reality.
What does this have to do with the ERP? Well, government bonds are perceived as safer investments relative to corporate bonds. This is because the government is viewed as less likely to default on its outstanding debt. It can simply print more money (or, “make printer go brrrr” in the contemporary vernacular) to pay off these loans, companies can’t. This means that government bond yield will be driven lower than corporate bond yield because they have lower bankruptcy risk. Hence, the gap in total return between equities and government bonds will be even greater than the gap between equities and corporate bonds. Again, this seems to be reflected in reality.
The gap between the average annual total return of long-dated government bonds vs. equities. Data ranges vary but all cover at least the range 1900-2015. Source: https://www.frbsf.org/economic-research/files/wp2017-25.pdf.
What would it take for this relationship to crumble? Some type of mix of very high interest rates and shocking equity total return might do it. Alternatively, the government could be viewed as a more risky investment than companies. Investors would demand higher yields to make the investment worthwhile, in this instance. This seems unlikely given the ability to print money to pay off debts, but, again, I wouldn’t rule it out entirely. Could you imagine a world in which Google is viewed as a safer investment than the US government? I certainly can.
The limits of understanding
Ultimately these two assets are completely different. Equities can fluctuate wildly in terms of yield, bonds (if held to maturity) do not. Equities can appreciate to theoretically infinite levels. Bonds, again, if held to maturity, can’t. Bondholders are more likely to get their money back in bankruptcy scenarios. Their payoff spaces are different, too (see above): both can go to 0 but only equities can go “to the moon”.
It seems the best we can hope for, as always, is to trust our reasoning and ensure it doesn’t conflict with historical observations. But at the same time acknowledge that we might be wrong and account for this scenario.