The third of the five critcial assumptions was that the retirement portfolio should be invested mostly in shares and real estate. This post will summarise the case for shares making up a significant part of the portfolio.
Leaving aside a recommended emergency fund which should be in a form which enables ready access without the risk of capital loss, the bulk of a retirement portfolio should be invested in assets that have the potential to produce a return which, after costs and taxes, is greater than the rate of inflation. A different way of looking at the problem is that a retirement portfolio should produce a stream of income which is not only adequate to fund retirement needs now but will continue to produce the same level of income in real (i.e. inflation adjusted) terms over the full duration of retirement (or at least for long enough that it makes no difference).
In an earlier post, I talked about life expectancies. If a retirement lasting for thirty years or more is anticipated, the compounding effect of even a relatively low rate of inflation will significantly erode the real value of a nest egg and it will do so at a time of life when options for remedial action are more limited than before retirement. Since the 1920s only two asset classes which are available to the average investor have generally satisfied this criteria over long periods of time: shares and property. Both produce streams of income (dividends and rent) which have increased over time. Certainly there have been periods of time, some lasting for several years, when shares and real estate have declined in nominal terms or in real terms or both. The great depression which started in 1929 is the most severe example during the 20th century. However, a look at any of the charts showing the returns of the various asset classes in the 20th century shows shares and property both providing meaningful real rates of return. Other asset classes such as bonds, deposits or even gold have failed to achieve this. One has to go back to the period from 1880-1920 to find a very prolonged period when bonds produced a better return than equites. Will things be the same in the future? I have no idea.
Are there risks in this strategy? Absolutely (and those risks need to be understood). However, with a safety margin built into the retirement budget, a contingency fund and at least three forms of "insurance" to fall back on, I will sleep a lot easier at night knowing that my hard earned savings and my standard of living were not being devalued by the mere passage of time. I will look at the three forms of "insurance" and the various asset classes in more detail in future posts.