Hi all, I was hoping someone can point me in the right direction. I am looking for some info on the historical return / st. deviation for different asset classes (i.e., investment grade fixed income, high yield fixed income, international equity, real estate, etc). Any help would be much appreciated. Thanks!
That’s a tall order. What kind of compensation are you offering?
hard to get for free. fastest way is to call a finance professor at a good MBA program
Ibbotson is the standard for this kind of stuff… Morningstar sells it, but it’s not cheap. If you are satisfied with just aggregate numbers over the long term and not specific period returns, you might look at: Triumph of the Optimists: 101 Years of Global Investment Returns: Elroy Dimson, Paul Marsh, Mike Staunton. About $100 at Amazon.com, or your local business library might have it.
good call. morningstar acquired ibottson a few years ago and that dude made coin. they will definitely have it.
- in Feb each year, Dimson Marsh & Staunton (London BS, published by ABN Amro) put out an annual series going back to 1900 for returns on 20 countries - - called “GLobal INvestment Returns Yearbook” - covers equities, bonds, st.devs, real/nominal, etc. Might find it on ebay as they sell out fast. Very useful on the face of it, but need to be careful about using such long time frames - because of structual shifts, non-stationarity, etc. Usually better to use more recent time frame, then test that underlying structural assumtions are still current
correction - 17 countries (should have checked before I posted…)
Canada - Equities [9.7%], Fixed Income [4.9%], Bills [5%] USA - Equities [10.3%], Fixed Income [4.7%], Bills [4.3%] EAFE - Equities [11.5%], Fixed Income [~ 6%], Bills [5.5%] These are longer term rates of return and the EAFE I sort of just mish mashed together. Hope this finds you well. Willy
Historical data will almost certainly understate variance in modern securities
Yes, that’s definitely true, for several reasons: 1) There may be rare tail events (Black Swans) that we haven’t observed in the historical data. The whole future is longer than the length of our known data, so black swans are certainly out there. It’s true that there might be outlying events in the historical data (Weimar Germany, for example), but it is more likely that there are more in the infinitely long future than in the limited past. 2) There is a structural change in how we share and manage risk, particularly with all these structured products. Although the total risk in an economy might be theoretically constant, I suspect that there are feedback and network effects that make it more risky. Actually, it probably means that you have more volatility clustering… periods of relative calm followed by periods of extreme volatility as people realize that they haven’t correctly estimated how much risk they are taking and are therefore taking on a lot of unpaid risk that they thought was actually paid.
I do a fair bit of research/writing/speaking on the Australian market and I have found that returns here are very close to log-normal (ie log returns are very close to normally distributed). Especially on the left tail (negative returns): monthly log returns since 1875 are almost perfectly normally distributed (eg around a 10y MA trend). But on the right tail (positive returns): there is consistent lepto-kurtosis and fat right tail - eg there are instances of 7 st.devs from normal (eg late 1987) - which with normal distribution would be a 1 in 65 billion year event using monthly data. So there have been more up-side black swans than downside - which is counter-intuitive. Of course that is history, the future is a different matter (but the patterns are consistent) Also, I have found that the patterns are consistent over very long periods - even despite huge structural changes in the economy, regulatory system, technology, risk management mechanisms, social changes, war, peace, depression, population explosions, inflation, etc, etc, etc - because ultimately the market is driven by human fear and greed cycles - and probably always will be…that at least is predictable! Other western markets would prbably show similar shape…