Modern Portfolio Theory Today

University of California, San Diego (UCSD) held a wonderful event October 16, celebrating the 25th anniversary of Dr. Harry Markowitz Nobel Prize in Economics for his seminal and lasting work on Modern Portfolio Theory. Among other activities, Dr. Markowitz still teaches at the University. And now, there will be an endowed Chair in his honor. If one looks at the body of work produced by Markowitz one could see other areas of research that would qualify for a Nobel. There have been many other awards in adjacent fields for his work. Needless to say, in spite of a changed world since his 1952 paper in the Journal of Finance, Modern Portfolio Theory remains at the core of investment activity today. And if one reads carefully through Dr. Markowitz’ papers, the kernels relating to almost all perceived advances and variations on MPT were anticipated and articulated in his work.  I was fortunate enough to be an attendee and a speaker at the event. I emphasize the attendee part as there were many major contributors to our industry in attendance. The opportunity to speak with many of them at what was a truly celebratory event, was worth the effort in preparing a ten-minute talk for such an intimidating audience. And, in this case I am not referring to the overflow crowd, but simply having Dr. Markowitz sitting there. It felt a bit like one’s orals as a final step toward getting that advanced degree. As one also knows, to paraphrase Winston Churchill, it takes hours to prepare a ten minute talk and ten minutes to prepare an hour talk.

Below are the prepared remarks for the event. A little history does provide a perspective.


Modern Portfolio Theory Today

It is an honor to be here this evening. I am not sure why I am up here. I am just a practitioner, not a researcher or an academic. I apply what has been developed. It just so happens, Modern Portfolio Theory (MPT) is and has been a key application. To talk about MPT today one does have to go back to the beginning to put into context what the state of the art was at that time and then discuss the context in which MPT is applied today. In 1952, when then, just plain 25-year-old Harry Markowitz published his paper in the Journal of Finance, the average daily volume on the exchanges was 1.9 million shares. Trading took place 6 days a week. The total market value, including what traded in the Over the Counter markets was about $181 Billion or about half of the reported GDP for that year at $368 Billion. A positive ratio, Warren Buffett would say. Over the next 10 years the equity markets more than doubled.  The average mutual fund holding period was about 9 years and stayed that way until the end of the 60’s.

When I first came to Wall Street as an analyst in 1968, the average daily volume was a little under 13 million shares with a couple of peak days around 21 million. The total market value of traded stocks including the OTC markets was a little over $1 trillion or slightly more than the GDP of $942 Billion. BTW, with some volatility the market ended a decade later at about the same level—actually 2% lower.  The trading took place for fewer than 5 days a week as the exchanges had to shut down for full days and occasional half days to process the trades. Help was on the way though, courtesy of Fairchild, Texas Instruments and Intel with the development of the integrated circuit and the microchip. Intel was actually founded in 1968, the same year I came into the business. Moore’s Law was in full swing by that time—processing speeds doubling every two years. So, doubling every two years, processing speeds today are about 8.4 million times faster than they were then. Yes, 8.4 million. And one does not have to put each line of code on an IBM punch card. This does lead to the ability to create some extraordinary very short-term algorithms around which to trade, based on an assumption that speed leads to greater certainty. I call it Rivkin’s Law. “The perceived utility (value) of information is inversely proportional to the time it takes to get it.” Notice, I say “perceived utility,”—not necessarily actual value. Not as punchy as Moore’s Law—just an observation on the big data phenomenon.

Bringing us into the present, the daily trading volume for all of the exchanges last Friday, October 9, was 6.8 Billion shares–520 times the volume in 1968.   The total market value of all equity securities is north of $21 Trillion vs. a GDP of $17.9 trillion—and we are only talking about the US. That ratio reminds a bit of 1968.  And, by the way, the average mutual fund holding period is down to less than 9 months. That says nothing about the average holding period away from the mutual funds. With the daily notional value of securities traded on the exchanges of around $250 Billion, the average holding period for all securities including the mutual funds is around 4 months. That means the average holding period away from the funds is about a day.  The ability to process the data, combined with the perceived reduction in transaction costs—notice I used the term “perceived,” once again–has led to different time frames to determine co-variance, the value of diversification, expected returns, and the likely variability of those returns. There is more information available to those making investment decisions—certainly more information that originally required human beings to develop and deliver in idiosyncratic and asymmetric ways. And the stability of all those variables comes into question, certainly over varying time frames and the correlations change as well.

In addition, particularly if we are talking about applying Modern Portfolio Theory today, I believe we are entering into an environment where even if one is actually making investments as opposed to trades, past performance may not be indicative of future results, particularly if one is relying on historical returns and risk measurements. “Past Performance is not Indicative of Future Results.” That wonderful statement that one finds on the bottom of almost every investment report. Although, this time I think it may be true. Much as it was true, beginning in the 1980’s when we moved from a long period of rising inflation and an accelerated move toward a service economy and a different mix of winners and losers compared to what we have, for the most part, experienced for the last 35 years—a very different regime. I was around in 1981, after living through more than a decade of flat returns for the overall market, when the Fed Funds rate peaked at 21.36%. Modern Portfolio Theory was being practiced at that time by a reasonable number of serious investors. Although, it unfortunately took the awarding of the Nobel Prize in 1990 before the general institutional investing public truly began putting those three words together. And, by the way, the average daily trading volume that year was 159 million shares, the market value of all US traded stocks was $3.4 Trillion, and the GDP was $5.6 Trillion.  The equity markets were up 4 ½ times in the subsequent decade.

This 30+–year period, since the peak in rates in the early ‘80’s, has been a very fortunate time to be involved in the investment business. A tail wind of generally falling interest rates, combined with a global economy changing dramatically, Moore’s Law at work, technological innovation, and global trade arbitrage of labor and capital, have benefited both the equity and the fixed income markets. We are coming to the other side of that trade. And to a time where one needs to heed the words of Dr. Markowitz from that first publication, “…of combining statistical techniques and the judgment of practical men” (and women—my add). And,  “…statistical computations should be used to arrive at a tentative set of expected returns and risks. Judgment should then be used in increasing or decreasing some of these [expectations] on the basis of factors or nuances not taken into account by the formal computations.”

I think we are entering another one of those periods where the judgments—the formation of relevant beliefs on the basis of observation–become more important, particularly at this stage of the transition. I have spent most of my investment life attempting to find those relevant beliefs and most active managers, away from those of us involved in allocations among asset classes, still continue that search. The element of risk as it relates to return remains an important part of that search.  Ultimately though, it all goes back to Modern Portfolio Theory, the Efficient Frontier and the portfolio optimization that comes from the direct application of the principles laid down by Dr. Markowitz and the Markowitz Model. Others have added bells and whistles–including the two Nobel winners who shared the economic prize in 1990. Others have pointed out additional considerations such as transaction costs and liquidity—important considerations—but all were already highlighted if one reads carefully through the papers of Dr. Markowitz. Those papers show a much broader set of contributions that include the application of MPT well beyond the traded securities investment industry, and other rational and Hume-like major intuitive observations and innovations in systems applications.

The investment industry specifically, but any other entity or activity that has to make some judgments about portfolio returns relative to the risks that may be incurred, must draw off of the principles developed by Dr. Markowitz. It is such an honor to be present on this occasion, much less a participant. Thank you, Harry Markowitz.  Thank you for this opportunity to share a few thoughts, and thank you all for joining in this celebration. As I said to my investment associates—a number of whom I see here this evening, if you haven’t had the opportunity before, being in the room with Harry Markowitz is one of those bucket list items for any investment professional. Pay attention and do take advantage of this opportunity. Thank you.

One thought on “Modern Portfolio Theory Today

  1. Pingback: Employment Triggers a Green Light for a Fed Rate Hike, but… | OUTSIDE THE BOXES: ALTEGRIS BLOG

Leave a Reply