Wednesday, August 31, 2011

Safe Savings Rates for the United Kingdom

As there never seems to be enough time in life to do everything on one's list, I'm still sitting on lots of research-related findings that I haven't had time to write up.  One of these issues is the matter for "safe savings rates" in countries other than the United States.  For as long as I've had the results for the U.S., I've also had results for 17 developed market countries using the data for 1900-2008 from Dimson, Marsh and Staunton, which I had used in my exploration of safe withdrawal rates for 17 countries. Aside from the general lack of enough time, one of the reasons I haven't written these results up yet is because I need to get around to doing the programming to calculate replacement rates as a percentage of net income (the income left after saving for retirement) rather than as a percentage of total gross income.  This isn't a major problem in the U.S., since the safe savings rates are not too high (though readers may beg to differ about the difficulty of saving this much), and I just explain that one reason why a 50% replacement rate may be more than it sounds is because you no longer have to save for retirement.  For instance, with a savings rate of 16.62%, this implies a net replacement rate of 50 / (1 - 0.1662) = 60% of net income.  That is, the 50% replacement rate actually corresponds to 60% of the income you were spending rather than saving prior to retirement.

For some countries, the safe savings rates end up being quite high (just like my paper on withdrawal rates showed that the SAFEMAX withdrawal rate is also quite low in some countries), which means it makes no sense to talk about this in terms of gross income.  One of these days I will do the programming to make these changes (the programming is not super straightforward because changing the savings rate means changing the net replacement rate and the two have to be balanced).

Fortunately, the United Kingdom is another place like the U.S. which does not have too high of safe savings rates. Philip Coggan, whose work I admire a great deal, recently wrote about safe savings rates for a UK audience, and in reading the article you can tell that he really wished he could be discussing UK numbers instead of US numbers.

With the data from 1900-2008 (compared to my US data since 1871), we can investigate retirements beginning between 1930 and 1979.  With the same assumptions, (60/40 asset allocation, 30 years of work, 30 years of retirement, constant real salary, and 50% replacement rate) the safe savings rate in the UK is 18.9%.  This compares to 16.6% for the U.S. since 1871 (with the worst case scenario in 1918 which is too early to be part of this data set), and 15.2% for the US from this dataset.



Here are the savings rates based on changing the asset allocation:

% stocks      0%     20%    40%    60%    80%    100%
savings rate  60.47  38.45  26.26  18.88  16.14  14.22

That is a fixed asset allocation held over a 60-year period.

And next, this figure combines information from Figure 1 and Figure 5 of the safe savings rate article, as it shows for the UK with a 60/40 asset allocation, and the 30 years of work and retirement and 50% replacement rate assumptions, the UK's maximum sustainable withdrawal rates in red, the savings rate needed to accumulate enough wealth to use the 4% rule in blue, and the lifetime-based savings rates in green. The safe savings rate was triggered by the first possible year to be considered (1930) and came close again in 1952.  I wouldn't be surprised if like the US, the safe savings rate in the UK would be higher if earlier years could also be included. Maximum sustainable withdrawal rates fell between 3.7% and 4.1% for the years 1900-1914. The withdrawal rate fell below 4% again in the years 1935-1937, which explains why the green line is higher than the blue line in those years.  Planning to save enough to use the 4% of rule would have been insufficient in those years, as more needed to be saved.  The green line does show a downward trend over time, and it will be interesting to see which direction it goes as we get data to consider more recent years.  Also, like the US, it is interesting that savers retiring in the year 2000 had the easiest time of anyone in history to reach a wealth accumulation goal.  They only needed a 9.8% savings rate to do it.  And that's pretty much safe savings rates in a nutshell for the UK.



At some point, I would like to more formally write up some results for all 17 developed market countries.  Before that happens, if anyone has a specific request for one of the countries, please let me know. 

Update: Just a quick note for anyone interested.  Among the 17 countries in this dataset, for my standard baseline assumptions, the US (15.22%) just edges out Canada (15.6%) to have the lowest safe savings rate of any country.  South Africa, Sweden, UK, and a GDP-weighted world index portfolio also  provide safe savings rates of under 20%  At the other extreme, and part of the reason why it doesn't make sense to talk about this in terms of gross income and I need to switch to net income, France has a safe savings rate of 102% and the safe savings rate in Japan is 240%.