Fecundity and Long-duration Retirements

I was just reading a paper by James Garland (The Fecundity Of Endowments And Long-Duration Trusts) and was struck by the affinities with early retirements. In fact, since I retired early, I myself have always had an eye on endowments and trusts as a model because -- while today at 61 if I were to abandon my DIY approach and plan on sinking a bunch of money into an annuity at age 80, my planning horizon is only maybe 19 years -- for early retirements the planning horizons can be really long, maybe up to 50 years or more.

While I do not 100% agree with what is implied by Garland's approach, since I do believe in "triangulation," this seems like another worthy way to think things through...it's worth a look. Me personally? I still think that the max "fecundity" of consumption portfolios with very long horizons (say 50 years to infinity) happens at a spend rate that is close to or above the horizon-adjusted expectation for multi-period geometric returns -- which depend on the horizon, realized returns, and volatility.  Maybe we are saying the same thing.  TBD.

Now is probably a good step-off point to define "fecundity." Most retirement math from the 20th century tends to focus on whatever spend rate and assumptions happen to "squeak by" over say 30 years given 20th century data. This has nothing to do with what Garland defines as fecundity.  Endowments and early retirees don't really play the game and do need to pay attention to what he calls fecundity. Garland defines this generally as

We use the term fecundity here to refer to a portfolio's long-term ability to generate spendable cash for its owner, because fecund means "fruitful or fertile," and cash withdrawals from a portfolio are effectively its fruit. Since the ultimate objective of all investment portfolios is to provide spendable cash over some period peculiar to the owner's particular circumstances, paying attention to the fecundity of long-duration portfolios would seem to make imminent good sense.

And later he goes into more detail to differentiate how we might evaluate or define things by horizon: 

The utility of capital depends upon the time horizon during which it will be consumed. For capital that will be consumed today, utility is a function of its current market value. For capital that will be consumed over a period of time — let us say a lifetime — utility is a function of future total returns. For capital that will never be consumed (or not consumed for a long time), utility is a function of fecundity. [emphasis added]

with a specific definition, in his terms, not mine, as (you'll have to read it see why he uses these boundaries):

• "the upper limit of the fecundity of stocks is their earnings yield ... 5-year or perhaps even 10-year averaging may be necessary to smooth earnings' hills and valleys ... history suggests that the earnings yield, rather than being the true measure of fecundity, instead marks at best its upper limit. The true fecundity of a diversified stock portfolio has historically been less than its earnings yield." 

• "the lower limit of the fecundity of stocks is their dividend yield." 

• "Logic suggests that endowment investors should be able to spend more than the dividends they receive from their equity investments, given that corporate retained earnings, i.e., earnings not paid out as dividends, are supposedly used to grow the business…If the fecundity of stocks is less than their current earnings, but more than their current dividends, then true fecundity lies somewhere in between...the fecundity of an equity portfolio — before expenses and (if applicable) taxes — lies somewhere between the earnings yield and the dividend yield of the portfolio. 

• "Endowment investment risk isn't a function of betas or Sharpe ratios or Value at Risk. Instead, the primary risk facing endowments is fecundity risk — i.e., the risk of a decline in the earnings of and dividends from the corporations in which they're invested. The futures of endowments and long-duration trusts [and retirements, btw, in this blogger's opinion] are inescapably bound to the futures of the American and world economies...The primary determinant of fecundity is economic health..."

[Note: he uses an arbitrary benchmark of 130% of the div yield for the fecund point. I'll use an entirely incorrect and arbitrary "split the two yields in half."]

That last bullet point is subtle but very important and shows why the most essential thing in retirement finance might not really be mathematics after all, but rather politics. We, however, are not touching politics today. Third rail and all. 

Now, let's take James at his word and assume that we have an observable set of market defined boundaries that need to be smoothed (we'll use 10 years here) and that also give us a range of spend rates that are tied to underlying economic fundamentals. Let's further assume that operating somewhere inside this range can maybe give us some hope of being able to consume our own "wodge" over a long enough time if our horizons are long and we have no lifetime income (annuities or pensions, or donations for that matter) to hedge out the longer term risk. Let's further assume for this post that we'll look, strictly for my lazy-blogging convenience, at an only-US-and-all-equity allocation (say S&P500). If we take the 10year SMA of the div yield and the earnings yield, this is what it looks like historically up to today....

Discussion

If we believe this and want to apply it, what is it telling us? I gather that it says that we can spend (today in constant terms?? or adaptive?  Not sure yet), if we have super-long horizons, somewhere between 2 and 5% with a quite arbitrary split-the-difference rate of around 3.5%. Can we calibrate this number? I think so.  I saw Wade Pfau recommend in 2017 or 2018 a sub 4% spend rate of something like 3.2 to 3.5%. It might have even been less; I remember 2.8% for some reason. Even ever-optimistic Michael Kitces, as recently as last week, suggested a 3.5% spend rate for early retirees if one is willing to adapt and take on more risk later. In any case, the current state of the market looks like it advises a wee bit of caution for long-horizon spenders.  That's me, btw.