“The fool doth think he is wise, but the wise man knows himself to be a fool.” ― William Shakespeare,
It’s been awhile so we are doing another edition of Blogger Wisdom this week on Abnormal Returns. As we have done in previous years we have asked an esteemed group of finance bloggers a series of (hopefully) provocative questions. Blogger answers are unedited and the author’s name, blog name and Twitter handles follow. We hope you enjoy these posts as much as we do putting them together.
Question: Assume you are advising a pension fund, endowment or foundation. What is a reasonable long-term expectation for real returns for a well-diversified portfolio? Support as you see fit. (Inspiration was this Vanguard piece.) https://personal.vanguard.com/pdf/s283.pdf (Answers in no particular order.)
The yield of a global portfolio is about as low as its ever been from a cyclically adjusted P/E, credit spread, and nominal interest rate standpoint, while the global economy is more likely to be in the later (than early) stages of the business cycle. Given that backdrop, I think a well diversified global portfolio will be lucky to earn 2-3% annualized real returns over the next 10 years, while institutions should have a plan in place if real returns are material lower than that.
I wasted way too much time over the last decade worrying about when the next shoe was going to drop. Had I instead allocated that time to identifying and investing in companies able to survive bad economic conditions and thrive in good conditions, I would be better off today. Yes, there is some hindsight bias in my response, but ultimately you have much better odds of finding a great company than guessing when and how the next crisis will occur.
My expectation is that stocks will deliver a 4% real average annual return over the next decade and a mix of high-quality corporate and government bonds will generate a little over 1%. That means a low-cost balanced portfolio might generate a real return that’s somewhat below 3%. I believe this forecast is built on entirely rational assumptions. I’m also highly confident it’ll be wrong, but I couldn’t possibly tell you in which direction.
I don’t know how we can know what is reasonable without predicting nominal rates and inflation. Suffice it to say everyone’s answer to this question in 1979 would have been different than it is today. My advice for the E&F crowd? Pay next to nothing for beta, and only pay up for potential alpha if it is truly differentiated, repeatable, and sustainable.
I’ll spitball based on what Jeremy Siegel said at our Democratize Quant Conference last week. (he’s way smarter than I am!). If you take operating earnings of S&P 500 you have roughly a 20% P/E, or a 5% yield. This is a real yield. Prof. Siegel doesn’t see a dramatic mean reversion in valuations (he mentioned that Shiller things it will drift to 15x at some point). So you have 5% real. If you look at 10yr TIPS you have less than 1% real, but let’s say 1% for easy math. 5%*60%+1%*40% = 3.4% expected real return. Not terrible, but if you can stomach equities and capture the 4%+ expected real risk premium — that ain’t bad.
I recently asked a group of over a hundred large asset owners whether they expected the average returns of those in the room to exceed 5% real over the next ten years. Perhaps five hands went up. I’m with them; I think it will be extremely hard to get to 5%, given the low rates on bonds and the high valuations on stocks and private equity. So I’ll say 4% real over the next ten years, with a bias lower than that. If inflation gets cranking, it will be less still, although higher rates and lower multiples will then allow for better returns beyond that time.
We’ve been in risk-on mode (and reaching-for-yield mode) for a long time. That doesn’t mean that it has to end now, or even produce the punk returns that I expect. But I think the odds favor it if you’re making plans. (Full disclosure: I’ve felt that way for quite awhile.)
Pensions and other perpetual pools of capital are essentially in a no-win situation. Not enough of them adjusted return expectations to account for ten years of historically low interest rates, they assumed interest rates in the fixed income portions of their portfolios that no longer exist. Reducing return expectations creates the political problem of increasing the extent to which they are underfunded. GMO is also on the record for expecting relatively low returns for the next ten years. Assuming Vanguard and GMO are correct about the next ten years there will still be a couple of year where the market is up a lot. If they can take defensive action such that they avoid the full brunt of one of the two or three terrible years, then their average return for the ten years will go up considerably. With no guarantees, reducing equity exposure any time the S&P 500 goes below its 200-day moving average provides the opportunity to miss some of those drawdowns.
Assuming a balanced 60-40 portfolio, it would be difficult to support anything more than low single-digit real returns. With the ten-year Treasury bond at close to 3%, and inflation around 2%, that’s roughly a 1% real return on 40% of the portfolio, while real equity returns can reasonably be expected to be anywhere from 3-5% at best. That gets you to around 3-4% for the portfolio. This is assuming, of course, that inflation stays relatively stable at 2%, which may be overly optimistic.
I’m not convinced by the argument that returns are going to be lower in the future, although I can see the logic behind it. But I think it’s sensible to err on the low side when setting expectations. A 4.5% real return for a 60:40 stock-and-bond portfolio is about right.
We all know the standard answer: stocks “always” return 7% to 10% per year. But while that might be true over a 20-30-year time horizon, the reality can be very different over shorter time horizons.
At today’s valuations, the S&P 500 is priced to actually lose 2%-3% per year over the next eight years. That estimate is based on historical CAPE valuations, which have limitations (including the failure to take into account differences in interest rates over time). So, let’s assume the CAPE is being unduly bearish given today’s yields and that stock returns end up being 5% better than the CAPE suggests. We’re still looking at returns of 2%-3%.
That’s roughly in line with with the yields you can achieve on a high-quality bond portfolio. So, core assets should return something in the ballpark of 2%-3% per year over the next 8-10 years. Overseas (and particularly emerging market) stocks might do significantly better than that, and commodities might enjoy a good decade starting at today’s prices. So, a diversified portfolio portfolio that included emerging-market stocks and commodities might post respectable returns. But a standard 60/40 portfolio is unlikely to return better than about 3% over the next 8-10 years.
If we’re actually talking long term then I’d anchor on the historic average. Yes, valuations are higher and interest rates lower today. But future stock declines and interest rate rises increases the return on reinvested gains. That has a big impact over time. It all comes down to how long term is defined. If it’s 20+ years, I’d worry less than most people.
That said, I’d plan on about 3% real as a return to use in planning, but with the emphasis on flexibility.
As an aside, I think it’s an interesting policy question if charitable orgs should be required to have distribution rates above “reasonable” expected long-term returns.
5-6%. put it in a no fee no load etf that replicates the S&P. Put some in a bond fund, some in high grade corporates. Avoid hedge funds and take 1% of the money and put it in PE/RealEstate/Venture to get alpha.
One thing I have gleaned from reading recent articles by Pew Trusts and P&I is that pension funds have looked farther afield beyond stocks and bonds in recent years. For many, the big push into “alternatives” (hedge funds, PE) and increasingly complex investment strategies has led to lower returns over the past 10 years, along with higher investment fees paid. The combination of greater expenses and lower longer-term returns will confound many pension funds and negatively impact their expected payouts to investors/employees.
Last year Bloomberg quoted Howard Marks of Oaktree Capital suggesting 5% as the reasonable return target for an “intelligently and prudently” managed pension fund. Skilled investment managers who can innovate and go beyond their consensus-hugging peers may find a path to higher returns with reasonable risk. As John Templeton once famously observed, “If you want to have better performance than the crowd, you must do things differently from the crowd”. I believe that applies to individuals as well as institutional investors.
3% real. I used mainstream assumptions for a combination of public equity, private equity, fixed income, and real assets. A heavier tilt toward less liquid and more volatility will buy you a bit higher returns, but not a lot.
This depends on the definition of long-term and the asset allocation of the portfolio.
I usually think of “long-term” as two or three decades. Nearly all the pensions, endowments, and foundations we work with at Plancorp make decisions under the assumption they will operate in perpetuity.
Within this context, I’m comfortable assuming an average real return of 5% for a global stock portfolio and 1% for a well-diversified bond portfolio over the next 30 years.
Around 4% to 5%. Roughly the 30-year Treasury plus 3% to 4%.
I am going to consider a “well diversified portfolio” a 50/50 global stock and USD hedged bond portfolio. I would expect long-term stock returns to be in a range of 5.25% real (15% standard deviation * 0.35 Sharpe ratio) to 4% real (2.2% yield + 1.8% real payout growth). I would expect long-term bond returns to be between 1.75% real (5% standard deviation * 0.35 Sharpe ratio) and 0% real return (current USD hedged real yield). Therefore, on the pessimistic side a “well diversified portfolio” would be between 2% real return (average of 4% and 0%) to an optimistic 3.5% real return (average of 5.25% and 1.75%).
It depends on your time horizon. 10 years – 3%/year. 20 years – 4%/year. 30 years+ – 5%/year. Infinite – you are joking, right? We are asking the wrong questions here, and assuming this is just a question of statistics, rather than a question of who will be able and willing to pay after a certain duration of time. We place too much trust in peace and political stability. There is less and less reason for average people in the US to support the present setup.
That is why I have started to write my Notes from an Unwelcome Future series. We are up against limits in the US as we have neglected orthodox economics and Christian ethics. Politics has become a zero –sum game, and economics is borrowing from the future. That will affect future returns, as does the current high valuations, low interest rates, and excess capacity.
It’s difficult to put together a high real return scenario given the many positives we’ve seen for the last 10+ years. The historical earnings and dividend yield has been about 7% on a rolling 10 year historical basis. But we’ve had pretty steady multiple expansion over the last 30+ years. If we assume that companies return their 7% earnings and dividend yield (perhaps optimistic) and we also assume that multiples don’t expand further (because they’re so historically high) then 7% is a safe starting point. Inflation has been 3% or so historically, but the good news for investors is that this has been trending down and given secular inflation headwinds, is likely to run closer to 2% than 3% in the future. So a 5% real return is a reasonable 10-year assumption.
I’ll take the “over.” Not over 5% necessarily, but over the Shiller/Hussman negative return for stocks.
This is impossible to present effectively in a few sentences, but here is the key point:
When expected inflation is low, the return from all assets is also low. Inflation expectations increase when economic growth (and profits) improve. Trying to forecast the next ten years is crazy. But if it is like almost any of the rolling ten-year periods in the past, it will include a period of strong growth, higher productivity, new businesses, and higher interest rates. What would we all think, in say 2015, if corporate earnings had doubled and the ten-year note was at 5%?
I hate to give a cop out answer here, but my response would vary significantly depending on what is meant by “long-term” and what is meant by “diversified portfolio.”
The classic “5% Real Return” target that is associated with institutional investment mandates seems quite daunting in the context of the next 5-7 years, but is much more realistic if your horizon allows you to look out 10, 15, or even 20 years.
As far as the portfolio itself, the more you can diversify the better your odds of success. With the ability to tolerate tracking error and relax the constraints relative to an arbitrary 60/40 type benchmark, there may be opportunity to deliver returns in the “5% Real” ballpark without having to load up on equity risk. Insurance-Linked Securities and Alternative Risk Premia are just two examples of diversifying elements that could add uncorrelated sources of return to traditional portfolios.
My expectation for a 60/40 Stock-Bond portfolio is ~3% real for the next decade. This comes from a 4% real equity return and a 1% real bond return expectation. I got this from reading a few articles that discussed the JPMorgan Capital Market Assumptions for 2018 (https://am.jpmorgan.com/gi/getdoc/1383498280832). JPM assumes a 3% nominal (1% real) U.S. Intermediate Treasury bond return with a ~3% equity risk premium, so there you have it. Obviously, I have no clue, but this seems to be a conservative estimation.
I don’t know what the right answer is because there is a lot of politics involved, but 7.5% is too high. I think 3-4% is reasonable. Low expectations that are easy to jump over are better than high expectations that have little chance of being reached.
Thanks to everyone for their time and effort. Stay tuned for a new Blogger Wisdom question tomorrow.