If Arnott is right, Encinitas' unfunded pension liabilities are a lot closer to the Stanford analysis of $154 million than the CalPERS/Blakespear estimate of $40 million.40% Bonds. Yield is 2% for the US aggregate bond market.
60% Stocks. Our base case is 5.4% for US stocks, but we think valuations are too high, so we trim this to 3.3% for the coming decade. Here’s our logic:
The yield is 2%.
Earnings growth over the past century has been 4.5%, of which 3.1% was inflation (real growth of 1.4% … far less than most people realize).
Inflation expectations are about 2%, so perhaps we should trim this forecast by 1.1%.
This gives us a base-case of 5.4%.
Valuation multiples are stretched, with the stock market priced at 25 times the 10-year average earnings, against a historical norm of 16.8x. If we’re back to historical norms in 10 years, that costs us another 4.2%. Since valuation multiples could (a) return to historical norms, or (b) remain at today’s lofty multiples, let’s split the difference, and trim our return expectations another 2.1%.
This gives us a likely outcome of 3.3% from stocks.
If our logic is sound, we earn 0.8% from our bonds (40% allocation x 2% return) and 2% to 3.2% from our stocks (60% x 3.3%, or 60% x 5.4%). Add up the return from stocks and the return from bonds, and we get 2.8% to 4% from our balanced portfolio.
Thursday, April 6, 2017
Renowned investor takes issue with rosy pension outlook endorsed by Blakespear
Widely respected investment manager Rob Arnott explains the math behind likely pension returns:
Subscribe to:
Post Comments (Atom)
I agree that the target returns are too high. This gives a simple way to look at it, but probably too simple.
ReplyDeleteHe discusses a reversion to historical mean on stock valuation (a negative) but not on bond yields (a positive). It's also unclear if he's ignored dividend yield in the equity portfolio, or if he's baked it in as a total return. If so, his total return on the stock portfolio is probably too conservative. Plenty of blue chips yielding 2-3% with very safe payout ratios (dividend payouts relative to earnings). Those dividends also tend to grow over time (on average, long term).
His point is correct, but his methods are flimsy. It's bad, but probably not that bad.
He does account for dividends ("The yield is 2%").
DeleteAs for bonds returning to higher yields, while that's a positive for investing new money, it doesn't help CalPERS' existing bond portfolio, which holds bonds with currently prevailing low yields (and which go down in value as yields go up). I'd be charitable though and say you could get 3% yield on a bond portfolio today, but that doesn't change the overall picture much.
Yes, but he gives a ten year horizon for normalization of equity valuations ("If we’re back to historical norms in 10 years"). If we gave the same horizon to bonds, plenty of the portfolio would likely roll over at normalized rates. The fluctuation in bond price shouldn't really matter, as I would imagine a pension fund is mostly holding to maturity anyway (or should be).
DeleteIf his stock projection is total return, the Next i think he's being too pessimistic. Inflation adjusted total return since 1950 has been 7% on average. He's knocking that down with a mean reversion on valuation, but the timing of that is unknown, and it's a one-time event. Pensions are a long game. Perpetual growth assumptions should not be predicated on one-time corrections.
Yes, but a lot of the historic returns on stocks are valuation expansion. I think he's right to use earnings growth rather than assume perpetually increasing valuations.
DeleteYes. I like using inflation adjusted earnings growth as a long term predictor. Real earnings growth rate for sp500 over the last 30 years is 6.05% (1986-2016).
Deletehttp://www.multpl.com/s-p-500-earnings/table
Valuation could revert to mean gently (a long period of equity prices underperforming relative to earnings growth), or disruptively (look out below). Probably the latter, unfortunately. The market could get cut in half without any change in the underlying earnings picture. That would put us at "normal."
Thanks- good link but I get 3.5% real earnings growth from those numbers:
Delete(89.89/31.92)^(1/30)-1
Are you compounding?
DeleteShould be straight line, no?
100x[(89.89-31.92)/31.92] gives the percent change for the whole period.
DeletePercent change/years= Annual growth rate.
You have to use compounding in this case.
DeleteTry growing the 31.92 at 6% for 30 years and you come up with a number way higher than 89.89.
i.e. CalPERS explicitly assumes that its investments will compound at 7% per year, which means much higher returns than earning a simple annual average of 7%.
DeleteI see what you are saying. Straight line gives 6% every year relative to year 1, not relative to the previous year.
DeleteIt's unintuitive to use compounding, since earnings in a given year have nothing to do with earnings from other years (unlike, say, interest compounding in a savings account).
But, you are correct.
Pension funds (and Social Security for that matter) have the appearance of a pyramid scheme. There's always a horizon where it could go broke and transfer the liability to our grandchildren when payouts continue to exceed money going in. However, in a long enough run it won't matter to our generation.
ReplyDelete"it won't matter to our generation"
DeleteNice.
In other news.... http://www.kpbs.org/news/2017/mar/31/north-county-transit-proposes-cut-bus-routes-rider/. Less trains and buses but our affordable housing numbers stay the same? What is going on here is not infill, it's sprawl. Sorry Lorraine El Rose, even if you could live here you would have no way to get to work.
ReplyDeleteWC,
ReplyDeleteYou're never going to get any satisfaction out of any of our elected officials on the pension deficit/issue. For an elected official, it would be like eating their own young. That said, really appreciate the info.
-MGJ
Pretty well sums it up, alright....
DeleteDitto to what 10:53 said.
ReplyDeleteBlakespear = another failed attempt at leadership.
ReplyDeleteBlakespare won't go against the unions that put her in office!
ReplyDeleteAnother Gaspar - just minus the high-heels.
DeleteYawn...... boring....
ReplyDeleteGo back to your Legos.
DeleteCan anyone tell me who was fired at the city? Thanks.
ReplyDeleteA while back it was reported that Manjeet Ranu and Mike Strong had left. I assume that's who people are talking about, but they probably technically resigned.
DeleteMash is the most important one, after too many years of subjecting residents to his insistence that every damn street in our city abide by his tunnel vision of property owners giving up a portion of their , 7' to 10' in order to provide bike lanes, sidewalks, etc, while saving nothing in the way like our treasured trees.
ReplyDeleteMany neighborhoods have had to band together to save their neighborhood character because of this cretin. Crest Drive, can you imagine all those trees being cut down to provide sidewalks? Wotan Dr., Saxony, Dewitt, the list is endless. Good freaking riddance on a much too long in coming move.
By the way, he was not fired. His job was eliminated.
What's the difference?
DeleteEliminating a position is the clean way to do it. The two positions eliminated are now held by Masih Maher and Gregg Shields. Their last day is May 19.
DeleteIt should have happened nine years ago. There are two more in Engineering with incredibly poor judgement and bad performance. Both are responsible for numerous F*(&^k ups on projects for Encinitas and should have been fired with these cuts.
DeleteThere is nothing wrong with firing for non-performance also. Many of these positions are not protected by the union like the two I am referring too.
Keep Cleaning.
Who are they?
DeleteThey could cut much deeper in the Planning Department. The environmental planner argues with residents and defends his bad work by saying that he had obviously made mistakes. That is what other people get fired for. Why is he still here?
ReplyDeleteIt's a start. Now let's eliminate a few more positions. Who do you think should be next?
ReplyDeleteNext layer in Planning. Two come to mind: one a boy, the other a girl. One caught multiple times lying to the Council, the other going along to get along - same diff.
Delete