SIR ELTON JOHN has a three-year farewell tour planned. This columnist has only a few weeks to go, before heading off to a new Economist beat. So it seems like a good idea to summarise some of the themes which have dominated this blog. 

To start, long-term investing. Here are a set of precepts every investor should remember.

  1. You can't start too early. Albert Einstein may not have said that compound interest is the eighth wonder of the world but it is a good motto to remember. Buttonwood started a pension plan for his daughters when they were three years old. Let us assume a return of 4% a year. That means a sum doubles in 18 years, quadruples in 36 and rises eightfold in 54. Looked at another way, say you have a set sum in mind for retirement. If you start saving at 20, you need to contribute only half as much money a month, as if you start at 30.
  2. Risk and reward are related, but don't think the latter is guaranteed. In financial theory, academics like Harry Markowitz and William Sharpe developed sophisticated explanations for the link between risk and return. This is where we get concepts such as the capital asset pricing model (CAPM) or beta, a security's riskiness relative to the market. But risk is measured in terms of short-term volatility. It is assumed, if you hold a risky asset long enough, you will eventually get rewarded. But this is not the case when you start from a high valuation—think of Japan in 1989 or the Nasdaq in 2000. Britain's FTSE 100 index is barely higher than it was at the end of 1999. A positive nominal return could have been earned from dividends but the real return this century from UK equities has been only 1.9%; real return from bonds 3.2%. Risk is not about volatility, it is about loss of capital. That is why investors should always have some money in cash or government bonds. 
  3. Long-term returns are likely to be lower from here. Even if equities do not perform as badly as in Japan since 1989, they are still likely to earn lower nominal returns from here. That is just maths. Short-term rates and long-term bond yields are low in both nominal and real terms. The return from equities is a "risk premium" on top of those rates. There is no plausible reason why the risk premium should be a lot higher today. The London Business School team of Dimson, Marsh and Staunton think it is currently 3.5%. Based on a return to mean valuations, GMO forecasts negative real returns for all equity markets bar the emerging ones (the same goes for bonds). US pension funds that think they are going to earn 7-8% are deluding themselves. 
  4. Charges are the financial equivalent of tapeworm. Say you invest $100,000 for 20 years and hope to earn 4% a year. There are two products available; one with an annual fee of 0.25%, and the other with a fee of 1%. How much more will the latter cost you? It is tempting to think the answer is small; it is only a difference of 0.75%. But the answer is $30,000 (see this SEC illustration). Of course, it is tempting to believe that the higher-charging product will deliver a higher return. But you don't know that; the one thing you know for certain are the charges. A recent FCA study showed that, net of fees, "more expensive funds have produced worse returns for the investor." Nor can you rely on funds that have done well in the past to do well in the future.
  5. Diversify globally. A lot of statistics about long-term performance are derived from America, which was the great economic success story of the 20th century. But this is an example of survivorship bias; back in 1900, people might have thought that Russia, or Argentina, would do as well or better. It is tempting for Americans to think that they don't need to invest abroad; most of the tech giants are based in the US. But the Japanese might have seen no need to invest outside their home market in the late 1980s, after its phenomenal post-war performance. The US market is more than half the MSCI World Index. It will not last. Diversifying protects the investor against currency risk and political mistakes. Economic power is shifting towards Asia (where it resided before 1500) and where more than half the global population lives.
  6. But don't specialise too much. The fashion today is to create thousands of different funds, covering ever smaller slices of the market. There has even been an ETF investing in ETF providers. Unless you are an investment professional who has researched the area extensively, you don't need this nonsense. Beware also of new investments that simply claim to be uncorrelated. That could just mean they don't rise in value when everything else does. The return from investing in equities is a share of profits; from bonds the risk-free rate plus credit risk. It is not at all clear what the return from investing in volatility should be (let alone cryptocurrencies). There may well be no expected return from them at all. So why buy them?