The very idea of an optimal portfolio has changed


Location Date: 
August 20, 2013
Content: 

The Times newspaper recently reported a surprising historic discovery deep in the vaults of a Berlin archive.  Recordings of British soldiers, actually prisoners of war, dating from 1916. Their strong regional accents define an age when transport and communications were minimal.  Almost no recordings exist of how ordinary citizens sounded then, so these are a revelation. Looking around our London office today, our team is, of course, British but also French, Russian, Polish, Canadian, Italian and American. It is doubtful any of us would even understand these people from a century ago, such are the gradual and inevitable societal changes. 

 

The OECD published a paper this week that attempts to look forward to the year 2060*.  It is mainly focused on long term economic and demographic trends with a view to the best policy choices.  While it’s impossible to picture the world 50 years hence, this report reminds us of the scale of change that is possible over multi-decade periods.  It also looks past the current post-crisis malaise and worries about QE tapering to a time when interest rates are back to historic norms in real terms.  Unsurprisingly China and India are expected to be a big part of the landscape then.  What is striking though is how much populations will age.  In many countries there will be nearly as many people over the age of 65 as under.  Looked at the other way round, only Israel, India and South Africa maintain or expand their proportion of working age population.  Realistically, labour participation rates have to rise in the coming decades to pay for healthcare and pensions for the elderly, not to mention rising interest payments on today’s long term public sector debt.

 

These time frames are highly relevant to interest rate cycles.  Looking at US data, it is likely that generationally low long-term interest rates bottomed earlier this year, just as the first baby-boomers started to retire.  The investment policy implications are clear.  A traditional stock-bond asset mix is not likely to perform as it has over the past 30 years in a persistently rising interest rate environment.  Indeed, today’s most sophisticated investors, including most sovereign wealth funds and large endowments, employ a hybrid combination of traditional and alternative strategies (hedge funds, private equity, etc.)  For the largest US endowments, this shift is impressive.  According to the National Association of College and University Business Offices, the average large US endowment (over $1 billion) is 61% allocated to alternative strategies compared to 29% a decade ago**.  Some prominent endowments are over 80% alternatives.

 


Source: NACUBO Endowment Study 2002 and NACUBO-Commonfund Study 2012 
 

Investment thinking has evolved dramatically over the years.  After the 1929 stock market crash, nobody wanted equities barring a few visionaries like John Maynard Keynes as Cambridge University bursar in the 1930’s and Warren Buffett in his 1950’s era partnership, which has evolved to today as Berkshire Hathaway.  The move to hybrid portfolios today is equally as important though alternatives are complex, time consuming and expensive to research properly.  It’s definitely worth the effort though.  Smaller endowments with lower alternative investment strategy exposure do not perform as well as larger ones.  This looks set to continue as rates move higher in the coming decades.  

 

Looking back, the British soldier’s voices from 1916 were predominantly rural, which serves to highlight another massive difference between then and now.  Today more than half the world lives in cities.  Not only have we completely changed how we live and work but the pace of change seems to be accelerating.  It is likely that the very idea of an optimal portfolio has changed too.  

Keith Tomlinson
Generation Asset Management

 

http://www.oecd.org/eco/outlook/lookingto2060.htm

** http://www.nacubo.org/Research/NACUBO-Commonfund_Study_of_Endowments/Public_NCSE_Tables.html