Does Modern Portfolio Theory Still Work?

By MBA, CFP® - Chief Executive Officer, Vernon C. Sumnicht on February 26, 2015

The following article by iSectors CEO, Vern Sumnicht, was originally published on in 2013. Vern wrote about the basic tenets of the original Modern Portfolio Theory (MPT) and discussed the challenges MPT faces in “the new” investing climate (of 2013 and beyond). The article also emphasizes how important the Post-MPT philosophy is to iSectors.

Does Modern Portfolio Theory Still Work?

The dramatic volatility and portfolio declines of the financial crisis led many advisors to blame modern portfolio theory. Yet after a careful look at its principles, it’s obvious that they are sound. Rather, it’s the application of these principles that is the challenge.

The booming stock market rally of recent years makes now a prudent time for advisors to revisit what we learned about modern portfolio theory and apply those lessons — before we find ourselves in the middle of another down cycle when mistakes snowball.

Before diving into the current challenges, let’s first quickly review modern portfolio theory’s core principles.

  • Investors are risk adverse.
  • Markets are efficient.
  • Asset allocation supersedes individual security selection and/or market timing.
  • Investing is for the long term.
  • Every level of risk has an optimal allocation that maximizes returns.
  • Diversification can eliminate specific security and industry risk. Correlation matters.

It is difficult to dispute these underlying principles, which are still as relevant as when they were conceived by Nobel Prize winner Harry Markowitz.

What challenges advisors today is not the core theories themselves, but their application in developing properly allocated portfolios for clients.


Advisors using mean variance optimization to determine asset allocation face four primary challenges: standard deviation, correlation of traditional asset classes, new economic factors and fees.

  1. Standard deviation is a poor measure of risk. The inherent flaw of using standard deviation as a measure of risk is that it considers upside and downside volatility as equally negative. This leads to the irrational assumption that investors are just as concerned with unexpected gains as they are with unexpected losses. As we know, investors are much more concerned with unexpected losses. Rather, advisors should use the possibility of losing money as a primary measure of risk. Doing so properly determines risk and uses a common-sense investing tenet: not losing money. This definition of risk is also supported by post-modern portfolio theory and behavioral finance research.
  2. Traditional asset classes are highly correlated. Advisors primarily use capitalization and style when applying modern portfolio theory principles. However, traditional cap-weighted asset classes’ large and small-cap growth and value correlation coefficients have begun to converge. A portfolio that uses only traditional asset classes and styles is no longer properly diversified and effectively reducing risk. Advisors must broaden their portfolios to include investments that are not similar. A more effective diversification approach would move beyond only styles to sectors. For example, research shows basic materials have little correlation to defense, chemicals and consumer durables. A portfolio today might increase exposure to this asset class. This same type of analysis can be applied across fixed income and alternative investments to improve real diversification.
  3. The traditional set of asset allocation factors is too narrow. To determine the optimal asset allocation, modern portfolio theory uses a mathematical approach called mean-variance optimization. That in turn uses only three basic factors to determine the risk-adjusted optimal asset allocation: risk, as measured by standard deviation, as well as expected return and correlation. Yet the economy, investment markets, investor utility and investment portfolios are all affected by more than three factors. Instead, asset allocation algorithms should consider relevant capital and economic factors when attempting to determine a portfolio’s optimal asset allocation or rebalancing decisions. Consider additional capital and economic factors — such as interest rates, inflation, GDP, unemployment, money supply, etc. — to determine asset allocation and periodic re-optimization.
  4. Management, trading and tax costs can be high. Modern portfolio theory never specified whether advisors should use active or passive funds to complete a portfolio allocation. But actively managed equity-based mutual funds without front-end loads charge about 1% to 1.5% each year to manage and operate the fund; additional transaction fees and taxes can also increase expenses. These costs are a severe drag on performance over time. Advisors should instead consider lower-cost vehicles, like ETFs or other index funds, can which offer investors a 2% to 4% lower total annual fee structure than actively managed funds. The savings can significantly improve client returns over time.

The principles of modern portfolio theory still hold true, and the theory works to achieve the objectives that many clients prioritize. But advisors must understand its limitations and find a practical application that increases its level of effectiveness in today’s markets.

Vern Sumnicht, MBA, CFP, is the CEO and founder of both Sumnicht & Associates and iSectors, an outsourced investment manager specializing in creating ETF-based investment allocation models.

Spread the love


Sign in to read more
Register to read more
By registering, you verify that you are a investment professional. If you have questions as to whether you should have access to the content on this site, please contact us.