Gyroscopic Investing

http://articles.businessinsider.com/201 ... ance-sheet

This is a guy who could appreciate the PP I imagine.

I'm watching this video of him too.

Craig would love this video! He has a similar view on correlations

http://www.youtube.com/watch?v=ze_zDhnc_ns

In the presentation Clive links to above there's a "capital preservation" portfolio, designed to withstand the extreme of the 1929 to 1932 market collapse.  This portfolio consists of

• 40% inflation linked bonds (global)
• 30% treasury bills (global)
• 20% treasury bonds (global)
• 10% gold

Ignoring the "global" part of this, it seems this could be implemented with TIP, SHY, TLT and gold (GLD).  According to etfreplay.com for the past 36 months this portfolio returns a total (not CAGR, not annnual, but total for the 3 years) of 30.7% with a volatility of 7.4% and max drawdown of 5.53%

The free version of etfreplay won't do it (too many funds) but a global version would be perhaps 20% each TIP and WIP, 15% each SHY and BWZ, 10% each TLT and BWX, and 10% gold.  I assume this would have about the same overall return with somewhat lower volatility and max drawdown.

By comparison 25% each VTI, SHY, TLT, and gold returns a total of 46.5% with a volatility of 9.4% and max drawdown of 8.65%.  The standard approach to reducing the volatility of the PP is to increase the cash allocation (keeping the other three balanced).  For example, 62.5% SHY, and 12.5% each VTI, TLT, and gold returns 26.3% with a volatility of 5.0% and max drawdown of 4.79%. 

Perhaps rather than half cash and half PP another approach would be half "capital preservation" and half PP.  Using the US funds for this would mean 20% TIP, 27.5% SHY, 22.5% TLT, 17.5% gold, and 12.5% VTI.  For the past 36 months this mix returns 38.6% with a volatility of 7.3% and max drawdown of 6.72%.  This mix might be more attractive to those who think interest rates are on the brink of exploding upward - although as Craig has repeatedly pointed out, how TIPs will react in a period of high inflation has not been tested.

melveyr wrote: I'm watching this video of him too.

Craig would love this video! He has a similar view on correlations

http://www.youtube.com/watch?v=ze_zDhnc_ns

He does dance very close to the issue. But I think he falls flat by actively trading. I really do think asset class correlations are bunk though. It's a classic case of applying the wrong tools to the job.

Saying asset X and Y are uncorrelated is just meaningless and also misleading. It provides no explanation why this relationship exists. Or, more importantly, why it should continue to exist. When people question this logic the supporters will push back that the economic data isn't strong. But then they'll happily accept the cop-out answer "Well correlations change over time."

So that's a better answer than pointing out that bonds are tied directly to interest rates which are tied directly to economic movements for example?

Their reasoning boils down to this:

"Umbrellas are open when it rains. Therefore, open umbrellas and rain are correlated!"

Well sometimes people use umbrellas on sunny days, too. Oops.

Browne's approach may not be perfect, but it's a heck of a lot better than what I see other people putting forth as investing theory. There is a lot of very bizarre stuff in the investing world.

rickb wrote: In the presentation Clive links to above there's a "capital preservation" portfolio, designed to withstand the extreme of the 1929 to 1932 market collapse.  This portfolio consists of

• 40% inflation linked bonds (global)
• 30% treasury bills (global)
• 20% treasury bonds (global)
• 10% gold

Inflation protected bonds are a very new product for US investors. There is no precedent in history as to what they may do if the US Dollar ceases to become a reserve currency for instance or under very bad inflation. In addition to this, they offer no protection under deflation as what happened in 1929-1932. So I'm not sure they are optimal for that situation. I'd want to hold cash and LT bonds if we get another 1929-1932 again. They are going to provide the strongest protection vs gold/stocks.

Also international inflation protected bonds are kind of pointless to hold unless you have exposure to inflation in those particular countries. Honestly the inflation in a place like Turkey (one of WIP's holdings) for example has almost no meaning to a US investor.

But even then I disagree with his advised portfolio because without stocks you can miss major gains if the economy recovers. Stocks are a primary driver of real gains in a portfolio and shouldn't be ignored no matter what we think could happen going forward.

Craigr "But even then I disagree with his advised portfolio because without stocks you can miss major gains if the economy recovers. Stocks are a primary driver of real gains in a portfolio and shouldn't be ignored no matter what we think could happen going forward."

To add to Craigr's point, the quickest ever doubling of the Dow was in the early 1930's in what turned out to be a dead cat bounce rally before the further fall to the 1933 low. Suppose you had decided (correctly) in 1930 that a depression was on the way. The best way to accumulate treasuries at that point would have been to hold stocks and then sell them to buy treasuries after the doubling. Obviously buying treasuries in 1929 would have been better still but in 1929 you wouldn't have predicted a depression unless you were some kind of soothsayer and then you probably would have been better off simply foreseeing winning lottery ticket numbers.

Clive wrote: I thought US TIPS, unlike our UK Inflation Gilts, repaid at par in negative inflation cases - and therefore would have gained in the 1929/32 period when multiple years of negative inflation figures occurred ? If so they can offer protection against inflation and deflation.

They pay par, but with the other assets in the portfolio falling in value (perhaps very badly) I don't think it's enough. Nominal bonds will beat them.

You also bring up another point why international inflation indexed bonds don't make much sense for a portfolio to protect against a 1929-1932 style deflation here. That is that they have different terms and conditions than US TIPS (often less favorable).

Clive wrote:

The Wall St Crash was largely just a reversal back out of the roaring 20's. An over-extension to the upside in the 20's followed by an over-extension to the downside to 1932.

Such events happen periodically. Anything that puts on gains of 20% annualised year after a year for a decade runs the risk of a subsequent large correction back down again. Gold more recently for example (since 2000) could turn out to be a more recent example. The shoe-shine boys buying stocks in the late 1920's could be akin to the gold ATM machines more recently popping up.

Thanks for the great chart, Clive. I think the important thing to take away is that a portfolio that was rebalanced would have been selling out of stocks the entire time and buying assets like bonds. Even with the relatively low bounce in the following years, the big savings was selling down stocks during the 1929 run-up.

That 1920s-1930s U.S. stock market chart looks a lot like the Japanese 1980s-1990s chart.