A Career Retrospective: My 35 Years in Finance

Morningstar Director of Research Paul Kaplan writes his last column before retirement, in which he reviews nearly five decades of education and work.

Paul D. Kaplan 29 June, 2023 | 2:57AM
Facebook Twitter LinkedIn

 Paul Kaplan Headshot

As I bring to a close a 35-year career in financial economics, I have been reflecting on key lessons that I learned first as a student, and then as working financial economist. I believe that is important to always be open to learn new things, no matter where one is on their journey. I have learned a lot over the past 45 years, and I hope to keep learning during my retirement years. So here goes.

Starting at the Beginning: Junior High School and High School

I grew up in the New York City borough of Queens and attended public schools starting in the first grade. In junior high school, I learned algebra and geometry, and so began a lifelong love of mathematics. In high school, I learned advanced algebra, trigonometry, and calculus. The skills that I leaned then, especially in algebra and calculus, I have used ever since.

I also got my first exposure to computer programming in junior high school and continued learning it throughout high school. As with mathematics, I came to love computer programming and continued to develop my skills throughout my schooling and career.

The greatest influence on me in high school was my economics teacher. In New York public high schools, a one semester class in economics was a mandatory part of the social studies curriculum. While I don’t remember exactly what she taught, it was this teacher that inspired me to take my first economics course in college.

Preparing Me for Life: College and Graduate School

I attended New York University (NYU) in lower Manhattan with the intention of studying mathematics and computer science. While I did continue to study those subjects, I took more and more economics classes to the point where I was able to triple major in mathematics, computer science, and economics. (There were very few required courses at NYU in those days, so triple majoring was possible).

The most important course that I took was in microeconomics where for the first time, I learned how economists build models from the ground up, starting with assumptions of rational behavior on the parts of consumers and firms. I also learned how economic models can have results that run contrary to many people’s intuition, and how economics be used to predict the unintended consequences of people’s actions and those of governments. Thus, economics is of great practical value.

After completing my bachelor’s degree at NYU, I went on the Northwestern University, where I completed by master’s and PhD degrees in economics. The classes that I took that gave me the skills that I would need in my professional career include microeconomics, econometrics, and the economics of uncertainty. Later I would see how the economics of uncertainty is a foundation of financial economics, and how economics and statistics are the foundations for empirical finance. One of the things that I learned in econometrics is that if two time series both have trends, they will be correlated, even if they are unrelated.

One of the classes that I took was a class in monetary economics. It was taught by a professor at the University of Chicago who was visiting Northwestern. He started the class by introducing some elements of financial economics and financial econometrics. He also recommended the book, A Random Walk Down Wall Street by Burton Malkiel. I read the book, and based on what the professor was teaching, I decided to write my PhD dissertation on financial economics. My dissertation consisted of two parts: (1) the presentation of an asset pricing model in which the issuance of government bonds can have an impact of all security prices (including stocks), and (2) the estimation of an econometric model of security returns. 

My Entry into the Workforce: Ibbotson Associates

My first job in the private sector was with a small firm in downtown Chicago called Ibbotson Associates. It was while I was at Ibbotson Associates that I earned the CFA designation.

Ibbotson Associates was founded and owned by Professor Roger Ibbotson of the Yale School of Management. (Ibbotson sold the firm to Morningstar in 2006, seven years after I had left and joined Morningstar.) The firm’s primary expertise was in asset allocation. It collected hundreds of time-series of returns on a wide variety of asset class indices. It sold this data with the option of including asset allocation software written by William F. Sharpe (who later won a share of the 1990 Nobel Memorial Prize in Economic Sciences.)

It was when I was at Ibbotson Associates that I learned about asset allocation and asset allocation software. The main component of an asset allocation system is a mean-variance optimizer based on the work of one of Sharpe’s co-laureates of the Nobel Prize, Harry M. Markowitz. When Windows 3.0 came out, we decided to replace Sharpe’s software with entirely new software that would work with a newly designed databased of asset classes indices. We decided to do this in an object-orient programming language that was specifically designed for writing Windows application. We decided that I would write

the optimizer, but to do that, I would have to learn the object-oriented programming language and figure out an optimization algorithm., I took a week-long class to learn the programming language. I figured out the algorithm from sample code (in a different programming language) in a 1987 book by Harry Markowitz, Mean-Variance Analysis in Portfolio Choice and Capital Markets. Since then, I’ve often written code using object-oriented programming.

For my first six years at Ibbotson Associates, I have the good fortune to work with Ibbotson’s first employee, Larry Siegel. (Larry is now Director of Research at the CFA Institute Research Foundation.) Larry is one of the best writers in finance and investments. My first experiences in writing for practitioners were coauthoring articles and book chapters with Larry. Larry and I also worked together in drafting expected return testimony for Roger Ibbotson in regulatory cases.

The Beginning of the Final Chapter: Morningstar, 24 Years Ago

After 11 years at Ibbotson Associates, I was recruited by Morningstar’s John Rekenthaler to work on developing the methodology behind Morningstar online advice system for people saving for retirement. From there, I became Morningstar’s director of research. In that role, I developed new versions of the equity style box, the star rating for funds, and Morningstar’s first generation of indices. I then when back to online advice, came back to central research, and relocated in London UK, to be the quantitative research director for Europe. In that role, I traveled throughout Europe, giving presentations, some of which were on asset allocation. I also gave presentations in Israel and the Far East.

Before leaving for London, I began working with Sam Savage of Stanford University. Sam is an expert in the use of Monte Carlo simulation in optimization problems. We developed an asset allocation optimization model that can work in conjunction with Monte Carlo simulation which we dubbed Markowitz 2.0.

While I was in London, I put together a set of articles (some going back to my time at Ibbotson Associates) and edited transcripts of panel discussions, debates, and interviews for my 2012 book Frontiers of Modern Asset Allocation. This book contains many of the things that I learned and figured out about specific asset classes, asset allocation, optimization (including Markowitz 2.0), and capital market history.

After two years in London, I relocated to Toronto where I have been ever since.

Notable Work from My Time at Morningstar Canada

While for the past number of years I have been part of the methodology team for Morningstar’s robo-advice offering for defined contribution plan participants, I also have had the opportunity to learn and write about a number of investment topics. Many of these topics I have written about in my Quant U column in Morningstar magazine. Below I discuss some of notable articles that I’ve published, you can find all of them here.

  • “Dollar-Cost Averaging: Truth and Fiction” (2019 Morningstar research study). Suppose that you just inherited a large amount of money. Should you invest it all at once or should invest it bit-by-bit over a period of time using a method called dollar-cost averaging? To many investors, intuitively, dollar-cost averaging is the better approach. Unfortunately, this is one of those instances where intuition leads investors astray. My Chicago-based colleague Maciej Kowara and I wrote this study to show why dollar-cost averaging is a myth. It turns out that the best time to investment money is the moment that you have it.
  • “Are Relative Performance Measures Useless?” (Journal of Investing, June 2019) In this article, Maciej Kowara and I take on one of the most common practices in investment management, evaluating a portfolio or fund against a benchmark index over a fixed period such 3, 5, or 10 years. Maciej came up a with different set of measures: how long can a fund with a manager with superior skill underperform a benchmark, and how long can a fund with a manager with negative skill outperform a benchmark. The answers to these questions turn out to be far longer than the fixed periods used to evaluate funds. Hence, the intuitive idea that superior manages should outperform their benchmarks over fixed periods of time (or conversely, that inferior managers should underperform) is wrong. Furthermore, the standard practice leads to unintended negative consequences in that it leads to funds with superior managers being overlooked and funds with inferior managers being held.
  • “Lies, Damn Lies, and Statistics” (Morningstar magazine, April/May 2018) The title of this article comes from a phrase that originated with Benjamin Disraeli and quoted by Mark Twain, “There are three kinds of lies: lies, damn lies, and statistics.” In this article, I discuss the problems with three types of statistical analysis that are often misused in the analysis of investment performance:
    1. Correlation. If two time series both trend upwards over time, they will be correlated, even if there is no economic relationship between them. To illustrate this, I plotted the level of the S&P/TSX Composite index and butter production in Brazil over time. The correlation is a whopping 87%. The way to correct for the common trend is to detrend both series by taking the percentage change of each. The correlation of the detrended series is an insignificant 5%.
    2. Back testing. In a back test, a strategy is simulated using historical data. The premise of back testing is that if a strategy would have done well in the past, it should do well in the future. The fallacy here is the future might turn out to be quite different than the past. Back tested results should never be used alone. The economic rationale for strategy needs to be sound.
    3. Relative Performance Measures. As Kowara and I discussed in “Are Relative Performance Measures Useless?”the common practice of selecting funds on the basis of performance relative to a benchmark over fixed periods can lead to funds with superior managers being overlooked and funds with inferior managers being held.

Books that I Coauthored

During my tenure at Morningstar’s Toronto office, there are two additional strands of research that I have engaged in, each of with resulted in a book published (or to be published) by the CFA Institute Research Foundation. In both these areas, I collaborated with my manager, Thomas Idzorek. These are:

Popularity: A Bridge Between Classical and Behavioral Finance (2018)

Authors: Roger G. Ibbotson, Thomas M. Idzorek, Paul D. Kaplan, James X. Xiong

In two published papers, Roger Ibbotson and Thomas Idzorek developed the concept of popularity as explanation for how nonpecuniary characteristics, such as brand recognition, reputation, and ESG characteristics impact security prices. If a characteristic is popular among investors, securities with that characteristic will be priced higher than they would be otherwise and expected returns will be lower. Conversely, if a characteristic is unpopular among investors, securities with that characteristic will be priced lower than they would be otherwise and expected returns will be higher.

I read Roger and Tom’s articles on popularity, and I was intrigued. I set about to form what their articles were missing, a formal model in which investors have nonpecuniary preferences. I call this model the Popularity Asset Pricing Model (PAPM). One evening in Chicago, I had dinner with Larry Siegel and he asked me if Roger, Tom, and I could put our work on popularity together into a CFA Institute Research Foundation monograph. Of course, I said yes and followed up with a proposal to the Research Foundation which was approved. In the monograph, we wanted to include some empirical work, so we asked James Xiong to contribute some statistical analysis (which he was already working on) on impact of brand, reputation, and other characteristics on security returns.

I find that the concept of popularity and the PAPM provide a very useful way approach the impact on nonpecuniary characteristics, especially ESG characteristics, on security returns. It is the hope of my coauthors and I that the PAPM will become a widely accepted asset pricing model.

Lifetime Financial Advice: A Personalized Optimal Multi-Level Approach (forthcoming 2023)

Authors: Thomas M. Idzorek, Paul D. Kaplan

We start this book by stating, “Lifecycle finance is the specialty in finance that focuses on the financial issues faced by individuals over the course of their lifetimes.” My first exposure to lifecycle finance was in graduate school. In fact, part of my PhD dissertation is based on it. When I started working at Ibbotson, I began further developing lifecycle models. I continued to do this when I joined Morningstar. For my column in Morningstar magazine, Quant U, I wrote a number of articles on lifecycle finance in which I developed models that include human capital, life insurance, annuities, uncertain returns, and dynamic spending rules. I also wrote about how to apply mean-variance optimization in the context of lifecycle models. All of these articles, taken together, form the basis of this book which I coauthored with Thomas Idzorek.

Tom and I hope that the approaches in our book on lifecycle finance will become widely accepted and acted on in the development of software for financial advisors. We believe that this will result on better outcomes for investors.

What’s Next

As I conclude the current chapter of life and am about to start a new one, I intend on continuing to learn new things and figure things out. Although this will be regarding different subject matter, I expect that the same principles will apply; namely be curious, keeping an open mind, and putting in the needed effort. 


Facebook Twitter LinkedIn

About Author

Paul D. Kaplan  Paul D. Kaplan, Ph.D., CFA, is director of research with Morningstar Canada.

© Copyright 2024 Morningstar, Inc. All rights reserved.

Terms of Use        Privacy Policy       Disclosures        Accessibility