Does Not Pass the Japan Test
After my piece yesterday where I ran a regression on Shiller’s data from 1871-1998, I got some comments pointing me to this piece at Clusterstock, citing the estimable Richard Bernstein. I will confirm what was suggested: after 10-year negative total returns, the returns are always positive over the next ten years in the US context, 1871-1998. The average is a ~7.5%/yr return.
But not so in Japan. We can talk about lost decades here, but in Japan the stockmarket has gone nowhere for over 20 years. Given some of the Japan-like remedies the US is pursuing, why should we expect returns that match the historical averages? Japan went through a period where debt levels were a very high multiple of GDP, like the US is at today.
When conditions are abnormal due to very high debt levels, one must factor that in, and not give credibility to statistics realized during periods with low debt levels.
December 10th, 200811:58 am at
Hello David,
I think the more apt analog for the US is the UK around 190 or Spain in 1700. Is it prudent to use the 20th century as a “bible” of all models given the fact that it was the American Century and unlikely to be repeated?
I think that the fiscal state of the US makes it far more like a banana republic then the Japan analog. Other than the reserve currency status of the US dollar, we would have already likely experienced a currency collapse ala Argentina.
Given this backdrop, I believe the debate should be centered on the stock market’s real return compared to an honest assessment of inflation. It think it is very possible for the market to generate 7-10% nominal returns over the next decade – the real question is how that stands up to any decline in the income and/or standard of living for citizens of the US.
December 10th, 20084:52 pm at
While I agree the US faces a “lost decade” scenario that may roughly resemble Japan, I see a few problems. Firstly, policy responses (whether or not they are a panacea [most likely not]) have been much more swift than the Japanese model. Much like a stroke patient, the first “minutes” are most critical for survival and recovery. Secondly, the Japanese market was so much more frothy. If you take the 2007 peak P/E at roughly 20x (which was skewed right due to the beginning of the negative bank earnings), that doesn’t even begin to compare to where the NIKKEI traded at the top. What are you thoughts? Really enjoy the blog btw (long time lurker).
December 10th, 20087:00 pm at
I think another aspect that breaks down the Japan analog, if I may go more micro, is the difference between US and Japanese businesses. US Corporations are more productive, less bureaucratic and have higher returns on equity. This surely counts in the US’s favor in regards to what lies ahead for US Corporations and in turn what lies ahead for the US Stock Market.
December 10th, 200810:32 pm at
Hi David,
I have a GREAT respect for your work. I was first introduced to you at Real Money a few years back and still remember your prescient call for GM’s eventual bankruptcy. A suppose the credit bubble kept them alive for longer than we all thought possible.
In any case, here’s my first piece of humor. Thanks again.
guidepostings.blogspot.com
Erik
December 11th, 20088:43 am at
In terms of “risk management,” is it more prudent to invest as if we are headed for a Japan-like scenario or as if we are headed toward a decade that meets historical averages?
Perhaps there is a third scenario to assume.
My observation, especially among discussions such as this that appear intelligent, is that knowledge does not by default translate into wisdom (good judgment).
In my humble opinion, proper risk management does not assume the worst possible or the best possible scenario and good judgment will certainly not make an individual cling to their opinions.
“Faced with the choice between changing one’s mind and proving there is no need to do so, almost everyone gets busy on the proof.” ~ John Kenneth Galbraith