Discover more from Economics Matters by Laurence Kotlikoff
Economics-Based vs Conventional Financial Planning
Is Conventional Planning Meeting an Appropriate Fiduciary Standard?
This column expands on an article published in 2018 in Forbes.
An enormous gulf separates conventional and economics-based financial planning. I am going to briefly lay out the differences, starting with these disclosures.
First, I am an economist and I treasure my profession’s century-long research on personal finance. This work has, by my count, contributed to a dozen Nobel Prizes in the field of Economics. Samuelson, Arrow, Merton, Modigliani, Sharpe, Markowitz, and Tobin are some of the heralded names in economics’ pantheon of financial geniuses. Each did seminal work on the economics of personal finance.
Second, through my software company I have developed the financial application, MaxiFi Planner, which delivers economics-based planning to households as well as professionals who are planning for their clients. Consequently, I have a commercial interest in my opinion. But academic economists, like myself, take an implicit oath to tell the truth, not only in reporting their research fundings, but in discussing economics with the public. I take that oath very seriously.
Pie-in-the-Sky versus Reality-Based Planning
Anyone who’s taken introductory economics knows that it starts with budget constraints — the proposition that we can only spend what we can afford. What we choose to buy among the set of things we can afford is our business. But we can’t spend beyond our means. That’s reality-based planning. But it’s not necessarily a grim reality. We can expand our budgets, and thus, spending ability, by raising our earnings (e.g., taking a better-paying job or switching careers or going back to school), working more, retiring later, finding ways to lower our taxes (e.g., Roth conversions, moving states, contributing more to our tax-sheltered retirement accounts), lowering our housing costs (e.g., downsizing), and maximizing our lifetime Social Security and other government benefits.
Conventional planning pretends the sky is the limit. How much do you want to spend in retirement? What’s your target? What are your goals? These are the first questions a household is asked by conventional planners and conventional planning software. Any alert economist would instantly answer — $1 trillion a year. Actually, make it $2 trillion. I just remembered more needs, goals, desires, and dreams.
Conventional planning takes this approach for a simple reason. It’s part of a well-honed formula to close a sale. First, bait the client with Target for whatever spending you’d like… I’m here to make your dreams come true... Can’t decide?... We’ll, apply the industry’s standard 80 percent retirement-income replacement ratio. No surprise, this ratio is miles too high for most households. No matter... It’s off to… Gee, making your target will be tough given how you are investing. But invest with me and I’ll raise the chance your plan will succeed; let’s check, uh… 91 percent of the time. Yes, there’s a fee, but your success rate is miles higher!
The economist would say:
Gee, I get to pay for a nine percent chance of losing my shirt? I’m trying to preserve my living standard, not gamble with it. And, gee, I was counting on spending $2 trillion a year. What happened? And gee, investing, as you suggest, in higher yield securities, comes with risk you aren’t disclosing. Yes, the probability of my spending the target you provided goes up. But risk isn’t free. The probability of my going broke earlier in life rises as well. That raises the number of years I’ll be starving. Thanks, but can you make a plan that’s realistic — one that starts with what I’ve got, not what I want?
Does Conventional Planning Meet an Appropriate Fiduciary Standard?
It requires advisors to put their clients’ interest above their own. They must follow the very best course of action, regardless of how it affects them personally or their income. A fiduciary’s advice must be the result of thorough and accurate analysis and they must execute it in the most efficient and cost-effective manner possible. It is important to avoid conflicts of interest as a fiduciary, so any potential conflicts must be clearly disclosed to their client.
Providing financial advice while simultaneously selling financial products isn’t precluded by the Advisors Act. Yet it raises its own red flag. But my concern, and yes, it’s coming from someone with an unavoidable major conflict of interest, is that conventional financial planning doesn’t meet the above description of appropriate fiduciary behavior. The method is at such odds with the economics of finance that it’s not, to my knowledge, taught in any top PhD programs in finance in any major university in the country.
Let me now describe economics’ approach to personal financial planning to clarify the differences. No approach is perfect, but one is grounded in standard economic theory and the other is not. At a minimum, households doing their own planning should compare the two approaches and planners should provide their clients with both analyses as well as explain their differences.
Economics’ Consumption-Smoothing Mantra
Economics-based financial planning originated with Yale economist Irving Fisher's seminal 1930 book, The Theory of Interest. This book laid out the basis for all of economics' personal financial theory, namely consumption smoothing. Consumption smoothing has two intuitive commandments — Don't eat all your eggs at once and don't put them all in one basket.
Consumption smoothing is grounded in physiology. We get satiated as we eat more at a given point in time and instinctively want to save for the future. Squirrels get this. That’s why they store nuts for the winter.
Here’s Fischer’s theory in, well, a nutshell — Spread your consumption capacity, i.e., your discretionary spending power over time, indeed — over your lifetime. This way you won't splurge today and starve tomorrow or do the opposite. Also, Spread your discretionary spending power over times — good times and bad times. This means don't invest in just one asset. Doing so will leave you spending high on the hog when that asset hits, and living on the street when it doesn't.
The same holds for insurance. Take homeowners insurance: if your house doesn't burn down, you consume more. But if it does, you consume less — a lot less. Buying insurance reduces your spending in the good state (no loss, but you're out the insurance premium), but raises it in the bad state (thanks to the insurance settlement). Hence, insurance, like portfolio diversification, smooths your spending over good and bad times.
All of modern finance is predicated on consumption smoothing. The desire to consumption-smooth is, by the way, intimately linked to the concept of risk aversion. Higher risk aversion translates, mathematically, into faster satiation. The more risk averse you are, the more you worry about the downside (consuming less) and the less you value the upside (consuming more), i.e., the more you want to smooth consumption.
Conventional versus Economics-Based Planning
As indicated, conventional planning doesn't calculate, as does economics-based planning, what a household should spend (on a discretionary basis) now, such that it can spend the same (maintain its living standard) in the future. Spending references discretionary spending — what the household can spend after meeting all its fixed/required/off-the-top expenses on housing, taxes, college tuition, etc. Conventional planning, instead, asks households to gamble on a desired spending goal that is likely far too high.
Of course, most conventional planners see the problem. They use the industry’s planning tools but have enough common sense and integrity to adapt them to try to produce a smooth plan. In particular, they will ask clients what they are currently spending and use that as the target. But most Americans are spending far too much. Hence, this approach is just a different form of highly dangerous target practice.
Sweating the Math
In addition to subtly helping pitch product, conventional planning is computationally trivial. Its Monte-Carlo simulations, which calculate a household’s “success” probability, require a modicum of high-school algebra. Economics-based planning, in contrast, requires advanced dynamic programming methods that jointly (internally and consistently) determine a household’s annual spending, taxes, and insurance needs subject to two constraints.
The first is lifetime budgeting. The household can’t spend in present value more than it has. This is also true along any path a household may experience in which their investment returns are stochastic. Second, annual spending can’t violate the household’s cash-flow borrowing constraints. Put differently, consumption smoothing won’t be perfect if you are cash-constrained. For example, you won’t be able to live as well in the short run as after your rich uncle dies and leaves you his fortune.
You might think that conventional planning could back into an affordable spending path by just adjusting pre- and post-retirement spending targets. But, due to cash-flow constraints, such trial and error takes literally forever. MaxiFi Planner does its consumption smoothing, subject to cash-flow constraints, in a half second with results precise to the dollar.
Economics-Based Versus Conventional Insurance Advice
Consumption smoothing has a clear prescription when it comes to the purchase of life insurance. Purchase enough life insurance to ensure survivors can afford the same living standard, as they would were no one to die prematurely. MaxiFi Planner’s half-second of consumption smoothing includes determining the household head’s and spouse/partner’s annual term life insurance needs. Conventional planning, in contrast, bases your insurance needs on your spending target. But if your spending target is too high, which will generally be true, you’ll be sold more life insurance than you need.
Conventional Portfolio Advice
What about portfolio advice? Conventional planning takes a household's annual pre-retirement saving as given and accumulates this saving plus initial wealth through retirement, based on rates of return drawn from assumed, time-varying portfolio holdings. After retirement, saving turns negative equalling the spending target. If the simulation leaves the household with wealth at some assumed end date, the simulation is counted as a success. The share of all such simulations that do not run out of money is conveyed to the household as the portfolio's success rate.
There are two huge problems with this method. First, if the targeted spending is wrong, which it surely is, the probability of success will be wrong, leading to the wrong investment advice. Second, conventional portfolio Monte Carlo simulations assume the household will spend, in retirement, the same amount (the target they set when they did their planning — potentially decades earlier) year after year regardless of whether their assets hit the jackpot or go down the tubes.
No one in their right mind behaves like this. Instead, as economics predicts, they adjust their spending up or down depending on how their portfolio performs. This assumption of blindly following some spending target through each-and-every retirement year, regardless of one's current level of assets, is hardly a reasonable basis for providing portfolio advice.
Moreover, as Wade Pfau, one of our nation’s leading authorities on retirement economics, and Massimo Young recently showed,
Probability-based retirement income strategies are highly sensitive to the capital market assumptions used in Monte Carlo analysis. Seemingly small changes in those assumptions can mean the difference between projecting a comfortable lifestyle and financial ruin.
This is an extremely damming critique of conventional planning given the huge dispersion that Pfau and Young report in Wall Street companies’ return distribution assumptions. Let me clarify their finding via an example.
Say planner A works for company X, which assumes the stock market will average a 5 percent real return with a 15 percent standard deviation going forward. And, say, planner B works for company Y, which assumes an 8 percent average real return and a 20 percent standard deviation. The two planners will, based on conventional Monte Carlo analysis, potentially provide dramatically different investment and other advice.
Company X might form its view of future market returns based on the last 20 years of data, whereas company Y might consider the last 50 years. Then there’s company Z that may use all available return data. From the client’s perspective, company’s X, Y, and Z may be indistinguishable. Indeed, they may be three of the largest financial players on Wall Street. But, depending on whose door the client walks through, she’ll receive, based on conventional planning’s methodology, dramatically different financial advice.
Economics-Based Portfolio and Spending Analysis
Economics has two approaches to handling investment risk. The first is called Certainty Equivalent Planning. This involves planning as if you’ll receive your plan’s assumed investment return for sure, but using a very conservative assumption about that return as well as all other uncertain inputs to adjust for risk.
Certainty equivalent planning is standard in finance. When we discount a stock’s future expected (average) dividend payout for risk we are engaged in certainty equivalence analysis. Indeed, all of asset pricing is predicated on this methodology.
The other approach is Stochastic Living Standard Risk-Reward Analysis. This approach uses Monte Carlo analysis to generate the range of living-standard trajectories a household can potentially experience if it follows a given investment and spending strategy.
As indicated, economics doesn’t countenance putting your living standard on autopilot. If your returns are low (or high), it says spend less (or more) right away. If you invest and spend more aggressively, your living standard trajectories will spread out and tilt downward. Hence, you’ll face both more upside and downside risk. You can limit that risk by investing less aggressively through time, spending less aggressively through time, or both.
By the way, adjusting one’s spending, in light of market outcomes, is also true of deterministic planning. Users will, after all, run their plans not once but, through time, adjusting their conservatively-determined spending to their updated lifetime resources, including their new asset values.
MaxiFi Planner does both certainty-equivalent deterministic planning as well as stochastic living-standard risk-reward analysis. Indeed, it has two forms of stochastic planning.
Full Risk Investing
The first is called Full Risk Investing. Here you tell the tool your current and future investment and spending strategy. An investment strategy might be, in the simplest case: hold a 50-50 stock-bond portfolio in all accounts through time. A spending strategy might be: spend as if I’ll always earn a 1 percent real return in the future. Armed with these inputs, MaxiFi Planner runs 500 Monte Carlo trajectories of your (or your client’s) living standard. It then shows the 5th, 25th, 50th, 75th, and 95th highest trajectories where the ranking is based on realized lifetime discretionary spending.
How sensitive is economics based Monte Carlo analysis to return assumptions — the issue raised by Pfau and Young? Not very as I’ll describe in a forthcoming newsletter. The reason is that living-standard Monte Carlo analysis is far less prone to sequence of return risk due to its annual spending adjustments.
MaxiFi’s second method of stochastic investment/spending analysis is called Upside Investing. Here, we are accommodating households or clients who are happy to entertain upside living standard risk but don’t want to take experience any downside living standard risk, whatsoever. Such households have, in the language of behavioral finance, habit preferences.
To accommodate these desires, MaxiFi asks users who run Upside Investing four questions:
How much do you have in the stock market?
How much and when will you add to your stock holdings?
When will you start withdrawing from the stock market?
When will you stop withdrawing from the stock market.
Next, MaxiFi builds a living standard floor for the household/client assuming every penny now in the market, and every penny that’s added, is entirely lost.
All assets not invested in stock are assumed to be invested in TIPS — Treasury Inflation Protected Securities — and earn the long-term real return on this safest of assets. Then, in its Upside Investing Monte Carlo simulations, MaxiFi gradually withdraws the client from the market per the client’s specifications. In so doing, MaxiFi invests all withdrawn funds in TIPS. And, whenever money is taken from the market and invested safely, MaxiFi raises the household’s/client’s living standard floor. Hence, the household/client experiences only upside living standard risk, with MaxiFi showing key percentile values of the upside trajectories.
Upside Investing allows risk averse households to have their cake and eat it too. They can sleep at night knowing their base living standard can never fall due to market performance. And, they can still be in the market and look forward to a higher future living standard based on the market’s historic excellent performance. This performance is generally so good that one need not invest a very large share in stocks to have a decent prospect of a significant upside. The key here is realizing that limiting living standard risk requires forming a joint investment and spending strategy.
Decades ago, the financial industry landed on a profitable planning methodology, which, in my admittedly highly conflicted view, violates a reasonable fiduciary standard. Conventional financial planning is at complete odds with economics and, indeed, common sense. As a result, it generates demonstrably inappropriate saving, insurance, and portfolio advice. My bottom line? Using the industry’s flawed methodology is unsafe at any speed whether you are planning for yourself or for your clients. But households and the clients of financial professionals should compare the two approaches and reach their own judgement.