Whether you believe that risk should be represented as a number or not, it is often accepted that, in equilibrium, the more volatile an asset’s return is expected to be, the higher the prospective return that should be demanded. So, for a higher volatility, an increasing premium is required above the return on cash. One of the (many) reasons that property receives an allocation in a multi-asset portfolio is the relative stability of unleveraged property returns. These are usually proxied by indices constructed from appraisals, which are less volatile than those constructed from market prices, where such indices exist.

Alongside this, property is also thought of as relatively illiquid. Liquidity relates to the speed at which you can turn an asset into cash (at or near market price) and this feature requires an additional risk premium to compensate investors for the time which their cash is “locked up”. This is because the longer the lock up period (and the more volatile the asset’s price during that period,) the greater is the risk that the price won’t be the same as its current price is thought to be. For real estate, the estimate of current price is a valuation, or appraisal and valuations are not prices. So, whilst real estate and other private markets are often thought of as “inherently illiquid” this might better be expressed as “inherently subject to price uncertainty”.

The illiquidity premium itself is the result of a number of things:

1. Delays to transactions because of the nature of property.

2. Delays due to the time taken to discover a price.

3. The willingness of the buyer / seller to trade at the discovered price. In certain circumstances, for example in rapidly rising markets, purchasers may be reluctant to buy at the market price and, similarly, in a downturn sellers may be reluctant to sell at the market price. This may cause a delay to trading or, trading may not take place at all. This is more of an “unwilling to trade at the discovered price” premium and it is questionable whether this behaviour should result in additional return for an investor at all.

Now, whilst it is oversimplifying things, let’s say that investors require 20bps of return for each unit of volatility; a valuation-based volatility of 10% would require a return premium of 200bps, or 2.0%, above the return on cash. If the price-based volatility, in a world where items 1 & 2 (and definitely 3) do not exist, is actually 15% then investors would require 300bps above the return on cash, a gap of 100bps of additional return. The gap between the two can be thought of as the illiquidity or “stability” premium.

In this example, when an investor says that they require, or think they will receive, an illiquidity premium of 100bps for holding property, perhaps what they are really saying is that they are prepared to hold property at valuation-based volatility, so long as the market compensates them for the real, price-based volatility. In this example, if cash returned 3% say, the investors would require a return of 6% for a volatility which is only commensurate with a 5% return.

In this way, real estate has an apparently low volatility and a relatively high return for that volatility. Plot these data points on a chart of risk (volatility) and return and it looks like a free lunch, but the risk element only includes the volatility element of risk whilst the expected return element includes the stability premium as well; a good example of having your cake and eating it. Put another way, if the stability in returns was real, no premium would be required for the illiquidity which that stability generates. Property investors aren’t being paid for illiquidity, they are being paid for a willingness to accept artificial price stability.

In amongst all of this, and to attend to the issue of oversimplification, there remains a real illiquidity premium but it is probably a small fraction of what it appears to be and it is being eroded by those seeking to speed up and standardise transactions.