For as long as we have records, one of the most persistent findings in financial markets has been that options are expensive. Implied volatility tends to exceed subsequently realized volatility, creating a variance risk premium (VRP) that sellers can harvest. This has been documented across markets, instruments, and time periods. It is also the economic foundation of numerous option-selling strategies and, more recently, an expanding collection of ETFs.

A recent paper by Ian Dew-Becker and Stefano Giglio, The Decline of the Variance Risk Premium: Evidence from Traded and Synthetic Options, challenges the idea that this premium is inevitable. They  find that the historically negative returns of index options have declined a lot. Over approximately the past fifteen years, returns to selling have been  basically zero.

This is an important finding. But I don’t think it implies that selling volatility has permanently stopped working.

The first point is that the variance risk premium is not a constant. It is a conditional phenomenon, driven by a complicated interaction of risk aversion, expectations, hedging demand, market structure, and the willingness of investors to supply insurance.

The familiar observation that implied volatility exceeds realized volatility is an unconditional average. It tells us very little about whether selling options today is attractive. There are times when options are extraordinarily expensive relative to their subsequent realized risk. There are also times when options are cheap. The premium can vary across strikes, maturities, underlying assets, and market regimes.

Risk premia are prices, and prices respond to supply and demand.

Consider the equity risk premium. Nobody expects it to be constant. Expected stock returns change as valuations, economic expectations, and risk appetite change. The same logic applies to volatility. When investors are frightened and willing to pay almost anything for protection, the variance risk premium can become very large. When investors are comfortable selling insurance and competition among sellers increases, it can shrink dramatically.

But Dew-Becker and Giglio provide an interesting additional explanation. They construct synthetic options using dynamically traded underlying assets, allowing them to investigate the role of investor preferences separately from some of the frictions specific to option markets.

Their synthetic options do not exhibit the strongly negative alphas historically observed in traded options. This suggests that the traditional explanation based entirely on investors’ aversion to downside risk is incomplete. Market structures and participants are always changed. Improvements in intermediation may explain part of the declining premium.

This is persuasive, but it doesn’t follow that the VRP has disappeared permanently.

There is a remarkably long history here. Scott Mixon’s (brilliant) 2009 study of nineteenth-century American option markets found that implied volatility generally exceeded realized volatility even in the 1870s.

But also, Mixon found that the gap between implied and realized volatility was substantially larger in the nineteenth century than in modern markets. This supports the idea that improvements in market structure and competition have progressively reduced the premium. However, improvements in market structure and competition have also progressively increased the ability to harvest the premium.

But it also suggests that the tendency to pay more for uncertain outcomes than their average realization would justify is deeply rooted in financial markets.

There is a psychological reason for this. Investors are risk averse. They don’t care only about returns (even if they say they do), but also when gains and losses occur. Making money during a market crisis is more valuable than making money during a boom. Options can win in bad states.

My expectation is that the VRP will return in more substantial form. Not because historical averages must reassert themselves, but because risk aversion and the constraints on those supplying insurance haven’t disappeared. History stretching back to the 1870s makes permanent elimination seem unlikely, although certainly not impossible.

This has an important implication for the growing industry of option-selling ETFs.

Many of these products systematically sell calls, puts, or other option structures regardless of whether the premium available is attractive. They effectively assume that an unconditional historical tendency constitutes a permanent trading edge.

But why should anyone sell insurance at a fixed schedule without considering its price? What other profitable business sells a product without considering price?

An insurer doesn’t offer the same premium when risk conditions change. A bond investor doesn’t accept the same credit spread in every environment. Yet an ETF that mechanically sells options every week or month is doing something remarkably similar.

This is the central weakness of permanent option-selling products. They eliminate the decision about whether the price of risk is adequate. Convenient implementation and predictable distributions are not substitutes for an investment edge.

The decline documented by Dew-Becker and Giglio is therefore less an obituary for volatility selling than a warning about treating an average historical return as a law of nature.

Disclaimer

This document does not constitute advice or a recommendation or offer to sell or a solicitation to deal in any security or financial product. It is provided for information purposes only and on the understanding that the recipient has sufficient knowledge and experience to be able to understand and make their own evaluation of the proposals and services described herein, any risks associated therewith and any related legal, tax, accounting, or other material considerations. To the extent that the reader has any questions regarding the applicability of any specific issue discussed above to their specific portfolio or situation, prospective investors are encouraged to contact HTAA or consult with the professional advisor of their choosing.

Except where otherwise indicated, the information contained in this article is based on matters as they exist as of the date of preparation of such material and not as of the date of distribution of any future date. Recipients should not rely on this material in making any future investment decision.