Photo by Nadir sYzYgY on Unsplash

Reality Defines Markets, Subjectivism Defiles Them — Part I

In 1990, Harry Markowitz won the Nobel Prize for his work on investment portfolio efficiency — an achievement he explicated in his 1959 book Portfolio Selection: Efficient Diversification. This work was later enhanced by William Sharpe’s Capital Asset Pricing Model (CAPM) and together, they became Modern Portfolio Theory (MPT). Essentially, MPT is a set of tools for maximizing long-term investment rewards at any level of measurable risk.

Not only did this improve fiduciary standards for institutional investors, but the MPT revolution delivered new wealth opportunity for individual investors, fueled the trajectory of the mutual fund industry and created new markets for brokerage firms and insurance companies. Of course, none of that would have been possible without the rise of big data, computing speed and the software needed to calculate efficiency for any mix of marketable stocks, bonds, commodities and cash.

As a widely used set of procedures that affect the lives of millions of people, it also deserves examination for its efficacy in solving problems. To that end, businessman Myles E. Mangram summarizes the premises of MPT in Global Journal of Business Research, “1.) Investors are rational (they seek to maximize returns while minimizing risk), 2.) Investors are only willing to accept higher amounts of risk if they are compensated by higher expected returns, 3.) Investors timely receive all pertinent information related to their investment decision, 4.) Investors can borrow or lend an unlimited amount of capital at a risk-free rate of interest, 5.) Markets are perfectly efficient, 6.) Markets do not include transaction costs or taxes, 7.) It is possible to select securities whose individual performance is independent of other portfolio investments.” [1]

For the purposes of this monograph, numbers 1 and 2 are nearly identical, 3 and 4 are nearly impossible and 6 and 7 are relatively unimportant. Regarding irrational investors with incomplete information (numbers 1, 2 and 3), that describes all of humanity, but it does not detract from the elegance of free markets. No one is omniscient, almost no one is perfectly rational, errors of knowledge are common — and that is not irrational.

What is irrational is to refuse to accept better ideas and discard old ones that are proven wrong. More to the point, it is irrational to ignore the power of decentralized, but incomplete information in the hands of billions of market participants who have skin in the game.

Yet, market critics dismiss the power of the price mechanism for discounting and transmitting all available information. To remedy that, the focus here will be on number 5 — the misunderstanding and misuse of “perfection” as it relates to a primary component of MPT — the market data. Specifically, the capital market assumptions (CMAs) that are to be integrated with an investor’s net worth and cash flow expectations for the purpose of making asset allocation decisions.

ON PERFECTION

Misapplications of “perfection” are everywhere — and a relevant example is quest for the “perfect competition” needed for market “equilibrium” in neoclassical economics. Not only does that ignore the role of the scientists and entrepreneurs who drive all competition, innovation and prosperity, but “Where this progression leads is unknown since one cannot reasonably divine the expansion of human knowledge or accurately forecast the capacity for relevant technological advancement.” [2] “Perfect” capital market assumptions are impossible for the same reason.

To believe otherwise is to give unearned recognition to economic central planning authorities — and that harms everyone. The essential difference between free markets and the State is mind vs. force. Ironically, it is “imperfect knowledge” on the part of government policy makers that undermines all attempts to make predictions and manage any division of labor economic system or add efficiencies. Freely determined prices are equilibrium — and they are transitory.

The ultimate danger is to define “perfection” that is inconsistent with the attributes of the subject. Regarding complex economic systems, they do not exist without voluntary production and trade for mutual profit — they are perfect when individual rights are protected and the initiation of force is banned. And because economies and markets are man-made systems, a similar approach to people and perfection is necessary. Perfection must be tied to the defining attributes of human life — free will, the absolute of reason and the ability to learn and grow.

The same is true and for the same reasons, with financial markets. Not only is it irrational to assign a purpose that is impossible by the nature of the subject, but with living entities and their creations it is immoral. Is a river not perfect if it floods after heavy rain? Is a bumblebee not perfect if it stings your face? Are men and women not perfect if they are actively learning and improving their ideas and decisions? Perfection must be based on the capacity of the subject — everything else is either faith or it is subjective.

And like the DNA signature of any living organism, there is one useful way to objectively identify a capital markets asset class — the most reliable data for their primary attributes. As Mangram attests, “it is reasonable to assume that only after a market or security has experienced a lengthy and proven record of healthy and consistent performance, under varying economic and political conditions, that historical market performance can be deemed a fair barometer of future market performance.” [3]

SUBJECTIVISM IN FINANCE

To be clear, “fair barometer” is meaningful, but it does not come anywhere close to “perfectly accurate prediction.” Nor does it need to. Nor should it. In support of the true nature and purpose of financial data, economist John Cochran (about the Nobel Prize winning work of Eugene Fama) wrote in the Chicago Booth Review, “The main prediction of Gene’s efficient-markets hypothesis is exactly that stock price movements are unpredictable! An informationally efficient market is not supposed to be clairvoyant.” [4] Precisely!

Cochrane is asserting the importance of being objective with historical market price information. That means all ideas and actions must be tied to the facts of reality. As a corollary, to discuss the possibility or impossibility of “perfect” capital market assumptions for predicting performance is a false premise. For professional advisors, it flirts with malpractice.

To summarize Mangram and Cochrane, “lengthy . . . varying conditions . . . healthy and consistent” market price data sets are best — and they “are not clairvoyant.” They are objective.

For each asset class, the proven long-term historical data is the most reliable gauge of potential market behavior. In terms of MPT, those defining attributes are median return, standard deviation, and correlation coefficient. Furthermore, none of them can singularly define a category. Each capital market assumption (CMA) is integrated with the others to more fully identify that asset class. In addition, they are correlated to the other categories for MPT’s diversification benefit.

Is all this common knowledge among the professional investment advice profession? Not so much. At least not according to the paper titled How Much Does Having the ‘Right” Capital Market Assumptions Matter in Retirement Planning? and published December 11, 2024, at www.kitces.com. In it, the co-founder and CIO of a financial software firm that specializes in cash flow and probability analysis, Justin Fitzpatrick, wrote, “However, for many advisors, using these assumptions isn’t always comfortable.” [5] Not comfortable? We’re talking about the primary ingredients of the investment strategy decision for their registered investment advisor (RIA) clients here.

Would you hire a physician who isn’t comfortable with lab results? An attorney who isn’t comfortable with legal precedents? A builder who isn’t comfortable with materials tolerances? A babysitter who isn’t comfortable with cognitive stages?

Where does this discomfort originate? The financial advice industry itself. Specifically, the beat the market value proposition of every market strategist and most fund managers. In turn, that is supported by the academy’s professors who churn out politicians that debase money, degrade prices, denigrate markets, and demonize profits. But I digress. Why are advisors uncomfortable with historically proven CMAs? Because they have been taught that their value is tied to analyzing markets, predicting the future and giving scant attention to price discovery and market efficiency. The alternative is subjectivity: to define reality without the benefit of empirical evidence, meaning “best guess” CMAs.

Fortunately, there is a solution to the market strategists who manufacture their own “forward-looking assumptions.” Also known as Monte Carlo simulations, it is probability analysis tied to an investor’s financial resources and cash flow expectations — and it changes everything. However, like any computer modeling system, the value of the output is directly correlated with the quality of the input. Think of it this way, how much sense does it make to use quantitative analysis as an objective replacement for subjective forecasts and then load subjective forecasts into analysis? None?

But this is precisely what is being recommended in the Kitces article by Fitzpatrick, for example,

1. “CMAs are most valuable when used as flexible tools rather than fixed forecasts.” [6] It is wrong to think of them as forecasts in the first place, but when they are considered flexible, everything becomes subjective and less valuable.
2. “Yet, even with the most accurate CMAs, financial advice rarely aligns flawlessly with reality.” [7] So what? They align quite well with their historical period, which is the most reliable gauge of future behavior.
3. “CMAs often prove wrong in retrospect, at least to some degree.” [8] Compared to what? Are the subjective ones better? It is wrong to ignore the range of potential returns in the most reliable CMAs.
4. “This weakness of CMAs for predicting the correct level of retirement spending is another argument for the use of adjustment-based ongoing planning.” [9] No, it is not. Flexibility with other variables is necessary, but CMAs must not be touched if the current market performance is reflected in the proven historical data.

On the other hand, if an advisor or their firm chooses to make recommendations based on “forward-looking” assumptions from their market research analysis, or use “conservative” assumptions, or use more recent history, that can be done with proper disclosure. But to use them in an ostensibly objective consulting process is a contradiction.

IDENTITY AND IDENTIFICATION

Invoking Aristotle’s law of identity, everything that exists has attributes that define it. And according to the law of causality, every existent behaves according to its nature. In other words, causality is identity in action. It is what it is!

But there is also a difference between the natural world and man-made phenomenon. Economically, it is impossible to predict and quantify the pace of innovation, “where this progression leads is unknown.” As a result, each CMA is a potential assumption error — there is no guarantee, at any point in time, that each asset class will behave precisely or even closely to the historical data. That does not make them imperfect, far from it. To repeat, the long-term historical data is the most reliable gauge of potential market behavior. To mess around with them only increases the potential assumption errors — and that’s a big deal.

In the framework of Modern Portfolio Theory, the three defining attributes of each asset class are,

1. Median expected return — “Expected return can be defined as the average of a probability distribution of possible returns.” [10]
2. Standard deviation of those returns — “Markowitz’ portfolio selection model makes the general assumption that investors make their investment decisions based on returns and the risk spread.” [11]
3. Correlations among each other — “MPT attempts to analyze the interrelationship between different investments. It utilizes statistical measures such as correlation to quantify the diversification effect.” [12]

One of the foundational precepts of economic science is supply and demand — and the same can be said about risk and potential reward for investing decisions. Higher risk investment options must have a higher potential return to be attractive. Conversely, lower risk options offer lower return potential. Quantitatively, this is obvious in the fact that every asset class carries a standard deviation that is positively correlated with its median return — the higher the return, the greater historical range.

So, what happens if advisors lower the median expected return of US stocks to conform with their firm’s research, or more recent history, or just to be “conservative.” Heck, “the concern may be that actual CMAs are too optimistic. No advisor wants to overestimate (overpromise) and underdeliver.” [13] To the objective advisor, the real concern should be the potential harm in lowering the standard deviation. Not only would that be necessary for an ostensibly lower return asset class, but it would also underestimate the likelihood of a severe market decline and force a more aggressive strategy to achieve the same goals. A perfect storm of “imperfect” assumptions.

In the Fitzpatrick paper on kitces.com, the focus is on “How much can I spend in retirement with perfect accuracy?” [14] As discussed, the premise is false. Perfect accuracy is irrelevant. To avoid pointless risk and lifestyle sacrifice are the important principles. Because of that, the illustration in the article is useless, “these periods with very different average returns would have supported the same retirement spending.” [15] Yes, but to an objective observer, the illustration only demonstrated the obvious: future returns are uncertain, a 30-year time frame is woefully insufficient — and sequence of return risk is a big deal.

OBJECTIVELY SPEAKING

Not only are the most reliable CMAs essential to an objective investment strategy decision, so are three other propositions: 1) Market performance is always uncertain, 2) The purpose of random simulations is to anticipate the possibility of extreme market events, 3) A contingency plan is needed in the event of #2 because of #1. In fact, entrepreneurs do that implicitly and their most reliable source of information is prices. Yet, investment industry experts who try to predict the successes and failures of entrepreneurs dismiss prices as irrational and markets as inefficient.

Regarding the Fitzpatrick paper, Michael Kitces editorializes by saying, “The absolute margin for error for CMAs would differ depending on the target portfolio. The higher the actual average return and standard deviation, the higher the margin for error in absolute terms.” [16] Not only does he adopt the false premise that reliable CMAs are not “perfect,” but he suggests that the median expected return is the defining characteristic of the asset allocation strategy and worse, overestimating the median return is more acceptable with higher risk portfolios. What could go wrong with that?

Like any other utopian project, it seems the thesis of the Kitces article is grounded in a postmodernist (nothing is certain, truth is personal) equilibrium target, “There’s nothing wrong with striving for better assumptions and a better model of the world.” [17] Yes, there is, if the assumptions are subjective. Existence exists. Models are approximations. When price information, investment risk exposure and objectively derived lifetime spending goals are replaced by a Platonist view of reality, the financial model may do more harm than good.

If a registered advisor was not comfortable with their capital market assumptions before the invalid statement “striving for not just accurate but extremely precise CMAs may not result in noticeably better retirement experiences for clients,” they will be less comfortable after reading it.

[1] Myles E. Mangram, Global Journal of Business Research, Volume 7, Number 1, 2013. Page 61.

[2] Ibid. Page 68.

[3] Ibid. Page 63.

[4] John Cochrane, Chicago Booth Review, May 20, 2014.

[5] Justin Fitzpatrick, www.kitces.com, December 11, 2014. Page 1.

[6] Ibid.

[7] Ibid.

[8] Ibid. Page 3.

[9] Ibid.

[10] Myles E. Mangram, Global Journal of Business Research, Volume 7, Number 1, 2013. Page 63.

[11] Ibid. Page 66.

[12] Ibid. Page 65.

[13] Justin Fitzpatrick, www.kitces.com, December 11, 2024. Page 8.

[14] Ibid. Page 4.

[15] Ibid.

[16] Ibid. Page 10

[17] Ibid. Page 12.