THE LONG-TERM RETURNS FROM COLLECTIBLES / THE ECONOMIST

Buttonwood

The long-term returns from collectibles

Investing in the finer things of life

BONDS, shares and Treasury bills are all very well, but in the end they are just pieces of paper. They are not assets you can hang on the wall or display to admiring neighbours. Many rich people like to invest their wealth in more tangible form; property, of course, but also collectibles such as art, fine wine and classic cars.

Is that wise? Elroy Dimson, Paul Marsh and Mike Staunton of the London Business School (LBS) have run the numbers for their annual analysis of the financial markets in the Credit Suisse global investment-returns yearbook. Some of these assets have done rather better than others (see chart). Fine wine delivered the best returns; surprising to cynics who might assume that, in the long run, the value of wine vanishes as it turns into vinegar. Really old wine often has historical resonance. A bottle of Chateau Lafite Rothschild from 1787 was sold for $156,450 in 1985 because it was thought to belong to Thomas Jefferson.

Estimating the returns from these assets, after costs, is tricky. Indices covering art or musical instruments are much less comprehensive than those covering shares. There may be an upward bias inherent in collectible returns, as successful works are more likely to survive.

Transaction costs, if valuables are sold at auction, may be 30-40%. But these are the kind of assets that tend to be held for many decades (and passed between generations) so the annual cost burden may compare reasonably with equities, which are traded much more frequently.

Then there are the costs of insurance. If people want to keep a Stradivarius at home, theft is a big risk; robbery with violins is a serious crime, after all.

But tax is a potential advantage for collectibles. Financial assets come with income streams that have historically been taxed at marginal rates of 40% or more. Art and stamps generate no income stream and incur tax only when they are sold. The academics calculate that, after tax, collectibles have generated higher returns than equities for British investors since 1900.

On top of that, investors may get an “emotional return” out of owning these assets, which may be as much a hobby as an investment. Anyone who has met an owner of a classic car will know they can display spaniel-like devotion to their vehicles.

What about the largest asset that many people hold—their home? The total value of global property was around $228trn at the end of 2016, against $170trn for equities and bonds. The academics are highly sceptical of a recent paper* that claimed housing has enjoyed equity-like real returns with less risk. They think this is an example of Twyman’s law: “If a statistic looks interesting or unusual, it is probably wrong.”

In terms of rental income, they say the study made “heroic” estimates of the effect of agency fees and voids (periods when the property is empty). When it comes to the level of house prices over the decades, the LBS academics say that a number of downward adjustments need to be made. The most significant is that the quality of the housing stock has improved. Over the past century homeowners have spent a great deal of money on extensions, central heating, indoor plumbing and so on. When all the adjustments have been made, the real return on housing has probably been less than on equities but more than on government bonds.

Perhaps the most surprising finding in the yearbook is that gold and silver have both done worse than cash and bonds over the past 118 years, despite high inflation during much of that period. In fact, gold performed best in real terms (although only as well as Treasury bills) when there was sharp deflation. Gold did substantially outperform T-bills during high-inflation periods, but this hedge comes at a long-term cost.

In the long run, equities have been the best-performing asset class, with a global real return of 5.2% since 1900. But that does not mean investors should assume those high returns will continue.

The prospective return on shares is equal to the real return on riskless assets (such as T-bills) plus a risk premium. That premium is now around 3.5% a year, the LBS trio think. As the real return on T-bills is currently negative, that suggests a real return on equities of around 3%.

The LBS academics made a similar forecast about low returns in 2000. The real return on shares since then has been 2.9%. If the professors are right again, more investors will be tempted by Bordeaux and Bugattis.


* The Rate of Return on Everything 1870-2015 by Òscar Jordà, Katharina Knoll, Dmitry Kuvshinov, Moritz Schularick and Alan Taylor