Forecasting is Hard…and a Fool’s Errand After Taxes

Forecasting is Hard…and a Fool’s Errand After Taxes

January 30, 2024

On the investor’s calendar, the turn of the year is a time for predictions.  Advisors like us and clients like you are deluged with “market outlooks” that offer perspectives on what’s next in the economy, the stock market and geopolitics.  Accompanying these are the investment equivalent of tabloid headlines…  Are interest rates peaking? Is the risk of a recession still on the table? What might 2024 hold for the stock market?  Who will win the election and what does it mean for your brokerage account?  Enquiring minds want to know.[i]

While fun cocktail party chatter, it turns out these annual forecasts are close to worthless helping you make better decisions.  To drive this point, we recreate an exhibit from Avantis Investors showed in the exhibit below the median stock market forecasts of Wall Street and the actual returns of the S & P 500.  The estimates ranged from 26% too low to 21% too high. 


Sources: Avantis Investors Field Guide December 2023 and YCharts. 

What about longer-term forecasts?  Many institutional investors – the big pensions and endowments – rely upon ten-year expectations from their consultants.  The thinking goes that the big annual gyrations will even out to arrive at some reasonable assumptions, right?

Wrong.  A recent paper surveyed 2012 assumptions from leading institutional consulting firms.[ii]  It found only one out of the fifteen assets classes surveyed had a return that fell within the range of expectations.  In other words, only one of the 15 asset class returns fell within the highest and lowest estimates.  Every other category was a collective whiff.  And these are the forecasts that drive trillions in investable assets!

But wait, it gets even worse.  The successful forecaster, in theory, would ride up an asset and exit right before it falls.  But, when we buy low and sell high, we get a big tax bill. 

How big?  For long-term capital gains (held over 12 months), the highest U.S. federal marginal tax rate is 20% plus the 3.8% net investment income tax, for a combined rate of 23.8%.  For California taxpayers, we also add the highest marginal state rate of 13.3%.  Adding it up, that’s a whopping 37.1% of gains interrupted by taxes.[iii]  But nobody talks about the impact of taxes on successful forecasting!

To get a sense of how big this tax delta could be, we recently completed research on what “Near Perfect” market timing would produce for a taxable investor.[iv]  We start with a 60% allocation to stocks and 40% to municipal bonds using applicable indexes.[v]  However, the strategy trades out of the stock market the year before a bear market begins and places the proceeds 100% in muni bonds.  A year after the bear market ends, the strategy trades back into a 60% position in the stock market. 

Unsurprisingly, this Near-Perfect strategy (which of course is preposterously impossible) would have produced stellar results.  Over thirty years, it would have outperformed a 60-40 portfolio by 1.1% per year.  Even better, you avoid the bear markets!  So the worst year would have lost only 8.5%, nearly a third of the loss of a 60-40 portfolio.

But here’s the rub.  Every time the Near-Perfect forecaster sold out of equities at the end of a bull market, their investors were handed a sizeable tax bill by both the IRS and the California Franchise Tax Board.  And each time the portfolio would reset lower after paying the taxes.  Before taxes, the Near-Perfect portfolio would have grown to $13.8 million from an initial $1 million over 30 years.  But after taxes, that figure is considerably smaller…$8 million by the end of 2022.  Still, an eight-fold increase is pretty good.  But, what if we skipped forecasting altogether and simply held the 60-40 blend?  We would have over $9.3 million, 15% higher wealth than the perfect forecaster!![vi]

Of course, perfect or even near-perfect market forecasting is folly.  But even if we can find a successful forecaster, odds are that they won’t add any value after taxes. 

Yet, market outlooks and their corresponding predictions are often at the center point of financial advisor and client discussions!  “Here’s what we see in the markets and here is how we’re allocating the portfolio” is a typical year-end discussion agenda for the wealth manager.  This strikes us as counterproductive at best and wealth eroding at worst. 

This analysis is a key reason why we emphasize “preparation over prediction.”  

Markets fluctuate, priorities change.  We are here to help. 

                            


[i] Sorry, every time John reads these scintillating questions he can’t help but think of his 1980s TV upbringing and commercials like this:  https://www.youtube.com/watch?v=AuE0a6uf3mM

[ii] The Accuracy and Use of Capital Market Assumptions by Mike Sebastian.   https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4184885

[iii] Portfolio composition and market prices can materially impact tax efficiency.  The actual tax efficiency of Flatrock portfolios will vary.  Clients are advised to discuss the tax impact of investments with their tax advisor.

[iv] We label “Near Perfect” as we could theoretically have constructed a forecasting methodology that had such clairvoyant timing down to the month, week, day, or hour.  Now that we be “Perfect”!

[v] We use the Russell 1000 Index for stocks and the Bloomberg Municipal Bond Index for bonds.  Performance shown includes the reinvestment of dividends and interest.  Dividends were assumed to be 100% qualified for tax purposes.  It is not possible to invest directly in an index. 

[vi] This comparison assumes no liquidation taxes for the 60-40 portfolio.  Remember - it hasn’t sold any of the stocks.  If it was liquidated, it would have a value a bit over $7 million.  That said, most investors will not pay full liquidation taxes on appreciated positions and instead rely upon other estate planning levers such as charitable giving, the step-up in basis at death, etc.  In the meantime, the portfolio produces a 15% higher dividend and income yield all else being equal.