We crush stock indexes, Yale claims

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
columbia
Posts: 1350
Joined: Tue Aug 27, 2013 5:30 am

Re: We crush stock indexes, Yale claims

Post by columbia » Sat Mar 09, 2019 11:19 am

One of those how to invest like Yale articles, that recommends a path completely unlike the Yale portfolio:
https://medium.com/datadriveninvestor/h ... d81a4918ff

Buttery Lobster
Posts: 21
Joined: Mon Aug 10, 2015 6:48 pm

Re: We crush stock indexes, Yale claims

Post by Buttery Lobster » Sat Mar 09, 2019 12:44 pm

This reminds me of an article I read last year about how Carthage College's endowment manager beat the returns of 90% of endowment funds over the past 10 years (and Harvard's by 1.8%/year).

https://www.bloomberg.com/news/articles ... ndex-funds

All with low-cost index funds. Pretty cool!
Image

User avatar
asset_chaos
Posts: 1486
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: We crush stock indexes, Yale claims

Post by asset_chaos » Sat Mar 09, 2019 3:47 pm

I'm all for passive over active, and hurray for Carthage College, but is it really that laudable that 90:10 beat 60:40 over a decade when stocks did really well?
Regards, | | Guy

iamlucky13
Posts: 1190
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: We crush stock indexes, Yale claims

Post by iamlucky13 » Mon Mar 11, 2019 2:52 am

Tanelorn wrote:
Sat Mar 09, 2019 10:57 am
Yale returned over 12% for their FY2018, exceeding a rough benchmark of 60/40 or 70/30 by 3% or so. Over 25% in hedge funds and under 10% in US stocks and bonds combined.
This kind of discussion is not helped by such generous use of the term "benchmark."

By the way, here's what they say in their report:
Yale has produced excellent long-term investment returns. Over the
ten-year period ending June 30, 2018, the Endowment earned an
annualized 7.4% return, net of fees, placing Yale in the top decile of
colleges and universities. Over the same period, domestic stocks
returned 10.2% and domestic bonds returned 3.7%.
Am I doing the math wrong, or would they have had roughly 8.25% annualized return if they'd stuck with a 70/30 portfolio?

Tanelorn
Posts: 1614
Joined: Thu May 01, 2014 9:35 pm

Re: We crush stock indexes, Yale claims

Post by Tanelorn » Mon Mar 11, 2019 7:13 am

iamlucky13 wrote:
Mon Mar 11, 2019 2:52 am
Tanelorn wrote:
Sat Mar 09, 2019 10:57 am
Yale returned over 12% for their FY2018, exceeding a rough benchmark of 60/40 or 70/30 by 3% or so. Over 25% in hedge funds and under 10% in US stocks and bonds combined.
This kind of discussion is not helped by such generous use of the term "benchmark."
I’m not sure what your issue is with it. They benchmark their hedge funds to bonds as these are “absolute return” funds (the kind that actually hedge, apparently, a rarity I know), which is 25% or so plus cash and bonds makes 30%. I’m lumping real estate and natural resources into stocks, which might not be a perfect fit since depending on what they hold those might be more like commodities than equities. Close enough for my purposes, call it 60/40 if you want to count those last two as an intermediate mix.

My goal with this rough exercise is to see if they’re beating something simple that an individual investor might hold with a roughly similar risk level and the answer for 2018 was “yes”. If that’s repeatable or not, who knows.
Am I doing the math wrong, or would they have had roughly 8.25% annualized return if they'd stuck with a 70/30 portfolio?
You did the math correctly (up to some minor rebalancing factor I’m not worrying about). Presumably they don’t want the risk of a 70/30 portfolio and view this much more diversified basket of alternatives as lower risk (and also somewhat lower returns).

Also, given their heavy current weight to foreign and emerging equities, which have underperformed domestic ones by a fair margin over recent years, that could easily be the difference. It’s not always true of course, but SP500 crushed everything for the last decade. Look at this chart for example vs VEU (world ex-US):

https://quotes.morningstar.com/chart/fu ... 2%3A955%7D

SPY outperformed by 5-10% annually!

international001
Posts: 668
Joined: Thu Feb 15, 2018 7:31 pm

Re: We crush stock indexes, Yale claims

Post by international001 » Mon Mar 11, 2019 8:02 am

bberris wrote:
Tue Apr 17, 2018 9:20 am
It is simply a math equation. The total outperformance of some equals the underperformance of the rest, minus costs. The net performance of all investors must equal the average performance minus fees. All this proves is that Yale got lucky. So as Dirty Harry says, "You've got to ask yourself one question: 'Do I feel lucky?'
Your logic is flawed. If we all do like Yale, we can all have returns above the average.

User avatar
Random Musings
Posts: 5349
Joined: Thu Feb 22, 2007 4:24 pm
Location: Pennsylvania

Re: We crush stock indexes, Yale claims

Post by Random Musings » Mon Mar 11, 2019 8:40 pm

We crush stock indexes?

Sounds like something Arnold S. would say.

RM
I figure the odds be fifty-fifty I just might have something to say. FZ

inbox788
Posts: 5927
Joined: Thu Mar 15, 2012 5:24 pm

Re: We crush stock indexes, Yale claims

Post by inbox788 » Tue Mar 12, 2019 11:58 am

asset_chaos wrote:
Sat Mar 09, 2019 3:47 pm
I'm all for passive over active, and hurray for Carthage College, but is it really that laudable that 90:10 beat 60:40 over a decade when stocks did really well?
My understanding is that Yale is active managed, including timing AA changes. I'm not sure why it's being compared to a 90:10 or 60:40 AA. As far as Carthage College, is the AA a static or actively managed parameter?
Tanelorn wrote:
Mon Mar 11, 2019 7:13 am
They benchmark their hedge funds to bonds as these are “absolute return” funds (the kind that actually hedge, apparently, a rarity I know), which is 25% or so plus cash and bonds makes 30%. I’m lumping real estate and natural resources into stocks, which might not be a perfect fit since depending on what they hold those might be more like commodities than equities. Close enough for my purposes, call it 60/40 if you want to count those last two as an intermediate mix.

My goal with this rough exercise is to see if they’re beating something simple that an individual investor might hold with a roughly similar risk level and the answer for 2018 was “yes”. If that’s repeatable or not, who knows.
It's clear the Yale basket beat most 60/40 funds and the SP500 in 2018. It's not clear to me that absolute return funds are like bonds or that they're similar risk levels. Using higher risk investments to get higher returns isn't too hard, just use a little leverage.

As far as commodities, they're not really expected to return more than inflation over the long run, so in some ways, maybe they're better compared to bonds. In any case, if Yale and the funds they invested in used commodities to beat equity returns, they successfully market timed (or got lucky).
https://www.betterment.com/resources/th ... mmodities/

inbox788
Posts: 5927
Joined: Thu Mar 15, 2012 5:24 pm

Re: We crush stock indexes, Yale claims

Post by inbox788 » Tue Mar 12, 2019 12:16 pm

international001 wrote:
Mon Mar 11, 2019 8:02 am
Your logic is flawed. If we all do like Yale, we can all have returns above the average.
I think it's the Yale comparison that's flawed. From their 2017 report linked in the article, on page 12:
Absolute Return
In July 1990, Yale became the first institutional investor to define absolute return strategies as a distinct asset class, beginning with a target allocation of 15.0%. Designed to provide significant diversification to the Endowment, absolute return investments are expected to generate high long-term real returns by exploiting market inefficiencies. The portfolio is invested in two broad categories: event-driven strategies and value-driven strategies. Event-driven strategies rely on a very specific corporate event, such as a merger, spin-off, or bankruptcy restructuring, to achieve a target price. Value-driven strategies involve hedged positions in assets or securities with prices that diverge from their underlying economic value. Today, the absolute return portfolio is targeted to be 25.0% of the Endowment, above the average educational institution’s allocation of 21.8% to such strategies. Absolute return strategies are expected to generate a real return of 4.8% with risk of 8.6%. The Barclays 9 to 12 Month Treasury Index serves as the portfolio benchmark.
Unlike traditional marketable securities, absolute return investments have historically provided returns largely independent of overall market moves. Over the past twenty years, the portfolio exceeded expectations, returning 8.9% per year with low correlation to domestic stock and bond markets.
They make the case that it should be an asset class, which if it's providing independent, low correlation to stock and bonds, I don't have issue, but then they compare the performance to treasury bonds. How and why they chose that specific benchmark is questionable. Their strategies may involve the use of bonds, but I suspect they're much higher risk and likely leveraged. The might as well compare to cash. IMO, it's akin to comparing Convertible Bonds to Total Bond Market or worse.

Another question to ask is whether Yale Absolute Return investments beat other Absolute Return funds.
Benchmarks for absolute return
https://www.ipe.com/benchmarks-for-abso ... 86.article
If a benchmark has a different risk profile from the absolute return fund an investor is comparing it to, how can he or she draw any conclusions about how much value is being added?

Comparing an absolute return fund with volatility of 10 per cent a year to a cash benchmark with near zero volatility is like trying to compare Formula 1 racing driver Michael Schumacher with the cyclist Lance Armstrong. Cash returns or inflation can therefore never act as appropriate benchmarks to measure skill. They are merely return targets or hurdle rates.
Absolute return funds need risk benchmarks
https://www.ft.com/content/e0b0b124-bff ... 144feab49a

User avatar
abuss368
Posts: 13414
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: We crush stock indexes, Yale claims

Post by abuss368 » Tue Mar 12, 2019 8:54 pm

Hard to compare with the high allocation Yale has to illiquid assets, private equity, and so forth.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Wed Mar 13, 2019 8:00 am

I'm well aware that this is the wrong place to attempt to reply to these comments. But in the event some folks are interested, here are a few answers to the proclamations and general questions in here, and every year, when Yale's numbers come out.

1. Hard to compare 70/30 (or any general mix) versus Yale in an attempt to match their risk profile. I get the idea of best approximation, but it's kind of missing the point. And I don't mean missing the point because the risk of financial assets isn't a great comp. What I mean is that isn't the risk they are diversifying away from.

An individual can have 80/20 (100/0) and then compare it over 10 years. What's missing is cash flows, and the timing of the flows. These endowments are primarily concerned about creating a portfolio that has <xx% change of losing more than% of it's spending power over any given xx year rolling period.

Feel free to try explaining that away with the last ten years, but that isn't maximizing performance while keeping their risks below what is acceptible -- all while having the right mix of assets where they can still meet a spending goal in a bad sell-off, while also having enough liquidity to avoid selling equities (and likely still able to add to dislocated assets from the hedge funds they can redeem from).

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Wed Mar 13, 2019 8:16 am

inbox788 wrote:
Tue Mar 12, 2019 11:58 am

My understanding is that Yale is active managed, including timing AA changes. I'm not sure why it's being compared to a 90:10 or 60:40 AA. As far as Carthage College, is the AA a static or actively managed parameter?

It's clear the Yale basket beat most 60/40 funds and the SP500 in 2018. It's not clear to me that absolute return funds are like bonds or that they're similar risk levels. Using higher risk investments to get higher returns isn't too hard, just use a little leverage.

As far as commodities, they're not really expected to return more than inflation over the long run, so in some ways, maybe they're better compared to bonds. In any case, if Yale and the funds they invested in used commodities to beat equity returns, they successfully market timed (or got lucky).
https://www.betterment.com/resources/th ... mmodities/
1. If you had to pick one, it's far closer to static. They aren't timing anything. Not to say endowments to jump on dislocations, but that is opportunistic based on a price and expected value... not tactical asset allocation

2. I say use 70/30 or 80/20. Flip the question and think about it like cost of capital, or where is the funding source. However, a lot of people on here are critical of certain benchmarks (like their AR benchmark), but only one person - maybe - pointed out how everyone is fine using the S&P500. Just helps emphasis how this exercise is so rooted in proving something. I say use the MSCI ACWI for stocks and use a 1y or 1-3yr Treasury index. Switch both benchmarks and it's a bit more relevant.

3. If you look at endowments (Yale too) 10 yr projections, they clearly assign far more return variance to absolute return compared to their fixed income. So they aren't saying it has the same risk profile. However, it's significantly more diversifying (many here may argue this, but first remind your self about the math which separates 'volatility reducing' vs 'diversification effect' statistically). So the nomenclature seems to be what has folks challenging the premise of the AR allocation.

4. For the Real Assets, you should assume this to be part of the equity allocation if you are looking to split everything into either stocks or bonds for comparison. It is going to be equity investments in real estate, oil/gas, and other real assets -- and to the extent it's a credit instrument, it will have equity like characteristics. (No top endowment has had commodity exposure in years, at least not the typical indexes available via ETF.)

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Wed Mar 13, 2019 8:34 am

inbox788 wrote:
Tue Mar 12, 2019 12:16 pm

I think it's the Yale comparison that's flawed. From their 2017 report linked in the article, on page 12:
Absolute Return
In July 1990, Yale became the first institutional investor to define absolute return strategies as a distinct asset class, beginning with a target allocation of 15.0%. Designed to provide significant diversification to the Endowment, absolute return investments are expected to generate high long-term real returns by exploiting market inefficiencies. The portfolio is invested in two broad categories: event-driven strategies and value-driven strategies. Event-driven strategies rely on a very specific corporate event, such as a merger, spin-off, or bankruptcy restructuring, to achieve a target price. Value-driven strategies involve hedged positions in assets or securities with prices that diverge from their underlying economic value. Today, the absolute return portfolio is targeted to be 25.0% of the Endowment, above the average educational institution’s allocation of 21.8% to such strategies. Absolute return strategies are expected to generate a real return of 4.8% with risk of 8.6%. The Barclays 9 to 12 Month Treasury Index serves as the portfolio benchmark.
Unlike traditional marketable securities, absolute return investments have historically provided returns largely independent of overall market moves. Over the past twenty years, the portfolio exceeded expectations, returning 8.9% per year with low correlation to domestic stock and bond markets.
They make the case that it should be an asset class, which if it's providing independent, low correlation to stock and bonds, I don't have issue, but then they compare the performance to treasury bonds. How and why they chose that specific benchmark is questionable. Their strategies may involve the use of bonds, but I suspect they're much higher risk and likely leveraged. The might as well compare to cash. IMO, it's akin to comparing Convertible Bonds to Total Bond Market or worse.

Another question to ask is whether Yale Absolute Return investments beat other Absolute Return funds.
Benchmarks for absolute return
https://www.ipe.com/benchmarks-for-abso ... 86.article
If a benchmark has a different risk profile from the absolute return fund an investor is comparing it to, how can he or she draw any conclusions about how much value is being added?

Comparing an absolute return fund with volatility of 10 per cent a year to a cash benchmark with near zero volatility is like trying to compare Formula 1 racing driver Michael Schumacher with the cyclist Lance Armstrong. Cash returns or inflation can therefore never act as appropriate benchmarks to measure skill. They are merely return targets or hurdle rates.
Absolute return funds need risk benchmarks
https://www.ft.com/content/e0b0b124-bff ... 144feab49a
Benchmarking absolute return funds (or hedge funds allocated in various ways endowments choose) is a constant challenge. The issue is, depending on why you are using the benchmark - or in other words, what question are you answering?

Let's say Yale runs a pretty tight absolute return portfolio where they aim to be market neutral (even if crudely). What that boils down to is 100% long one asset(s) and 100% other asset(s). Because you park the short rebate into a treasury account, the exposure of the absolute return is actually exactly that of the benchmark, T-bills.

What seems to be conveniently overlooked is that Yale and most other top endowments also have a secondary list of benchmarks which are peer groups. Seems like this is what you were seeking in the first place, and they do in fact compare to peers across all asset classes (and to far more granular degree than will ever be in a public report). The policy benchmark is specifically built with only investible indices to provide the best effort expressing what the underlying risk factors are. (So, ex-ante, any type of peer group obviously provides absolutely zero value in such a utility function.)

Finally, Yale actually specifically points out the problem with bucketed their portfolio by 'asset class.' They write it every year in the opening section, something to the effect of the flaw in defining asset classes where no clear definition exists. Seems pretty fair, right?

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Wed Mar 13, 2019 9:10 am

Valuethinker wrote:
Mon Apr 16, 2018 11:58 am
Boglegrappler wrote:
Mon Apr 16, 2018 11:14 am
So Yale could be overstating the value of its funds, where there are unrealized investments in those funds. However eventually the value will converge on the true cash value. Only if they keep putting more money into funds, at overstated valuations for unrealized investments, will this lead to an unrealistic overstatement of value - in equilibrium with cash in to the portfolio = cash out, the valuations will be on average correct.
I found the rule change that has resulted in the reporting change that I noted. If you look at their 2014 financial statements, you can see that about 20B of 29B in assets were classified as "level 3", meaning that the valuation of those assets involved a great deal of estimation. Beginning in 2015, only a small subset of assets was subject to the level 1,2,3, classification, while the vast remainder is simply presented as net asset value (NAV). This resulted from accounting standards update 2015-07. I can't say that I'm up to date on these issues, but its clear from the before and after that a potential red flag was removed for three quarters of their assets. (Again, its important to note that this isn't just Yale, its everybody who runs large investment funds that invest in illiquid investments.)

My fear is that it puts the valuation issues into the hands of the general partners who run the funds and their auditors and appraisers, and in a way that could lead to some surprises down the road. We can only hope that does not turn out to be the case.
Ahh yes. Thank you.

If subsequently the GPs have been systematically overstating values, then things will look pretty well since 2014, and that could come home to roost.

There's an issue with doing that, if you are GP. There exist Fund Secondary buyers (Coller Capital in Europe, I know Paul Capital Partners in USA) who buy units in funds e.g. when a pension fund or endowment needs to generate liquidity quickly before the actual windup of the fund. I am guessing that funds change hands more or less at NAV per unit.

If there is significant overstatement then the sellers are winning, and the secondary funds buying are getting done over. And that will lead to litigation against the GPs & their auditors.

So the risk is there, and it's real, but the GPs cannot operate with a completely free hand. On the other hand, the Yale team itself might have an incentive not to query NAVs too hard, depending on how their compensation is structured. It was annual bonuses that made creating and flogging CDOs so lucrative and attractive for investment bankers, to the detriment of their firms and their clients.

I can see a risk there and it's hard to quantify it. You see people like the Canadian public pension funds going all in for direct PE investing (ie not through a fund) and there's an incentive (their own bonus schemes) to hold at unrealistic values. And that's significant money (billions).

If we could get a sense at the premium/ discount to NAV Fund Secondary buyers are paying then that might give us a sense of the risk.
GPs aren't the ones just making up a valuation. The only way they'd be liable for anything (in the context of your point) would be for fraud. Secondary funds buy interests above and below NAV all the time (that's kind of the point of their strategy... they may think a premium to the most recent mark will still give them the expected return they need). Secondary transactions do not involve the GP outside from allowing the sale.

As for compensation for the LP investor via an endowment etc., they don't really control any of the variables that would get them paid (or not) based on bloated private unrealized fund valuations. So I don't really understand the parallel you draw to bankers selling product for commissions. This simply isn't have private assets get marked.

Great point on the compensation conflict though. They all know, at this stage in a cycle, there is a good change much of unrealized value can vaporize... Most smart institutions (like Yale) pay their private markets analysts a bonus that is based on a multiple year rolling cycle, so the big losses ultimately have to be made back up. Btw, the fund managers aren't getting their profit share on the unrealized money either, so they actually are incentivized to realize gains.

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Wed Mar 13, 2019 6:52 pm

Tyler9000 wrote:
Mon Apr 16, 2018 9:54 pm
CULater wrote:
Sun Apr 15, 2018 2:37 pm
Maybe passive doesn't always beat active? At least for institutional investors.

https://www.institutionalinvestor.com/a ... ale-claims
Thanks for the link. Very interesting.

I started writing a reply, but it quickly got out of hand and I realized I had way more to say than could be summarized in a simple message board post. So I did a full write-up instead: Things We Can All Learn From The Yale vs. Buffett Debate

I hope you find it interesting. Your topic certainly took up way more of my day than you realize. :wink:
This was a pretty interesting read, and unlike many who write about the topic, it seems you did the best with what you have to present fairly.

A few questions, and a couple comments to add some color:
1. When you are comparing drawdowns, are you simply linking annual returns? I'm assuming that's all you have, and if that's the case, your readers need to understand that those intra-period variance and drawdowns can be very different than what's stated. This is critically important due to the timing of the endowment's annual spend (which is pretty darn big for them).
2. To do a fair comparison against a passive dummy portfolio, it needs to be assessed on a dollar-weighted basis (but, no one has that data). However, because the Yale portfolio is has less return variance, it's likely it would further favor the endowment here.
3. Importantly, you mentioned that Swensen was not at Yale dating back to the beginning of your analysis. But even more prescient to this specific topic, the 20 year look-back is actually probably closer to an honest time period of cover only the 'Yale Model' performance. Of course he'd built out a lot of the AR, PE, VC before 1998, but not much before that - and not nearly to the extent it became by 2000.
4. 60/40 had just benefited from a secular bond market where the 40% bonds almost without certainty produced returns that won't be matched in the coming 20 years. Also, US equity far out-paced International equity during this period which makes the 60/40 past performance even less relevant as we move forward. (Notably, Yale doesn't own much of the higher returning bonds in the bond index... and they have been overweight EM and DM compared to US stocks, which has been a headwind).

Some less tangible points:
1. When comparing Yale's (and other top 20 endowments) equity performance to that of the asset class specific index it's completely fair to assess the numbers at face value. However, if you are attempting to extrapolate whether the endowment team has can repeatedly identify skilled stock pickers, you'd need a good deal more information (as I am sure you're aware). What I can share is that the aggregate exposure of their investments within each 'asset class' segment is very different from that of the passive index. With concentrated portfolios, and high levels of information ratio, yet still keeping volatility below index is impressive... And back to the IRRs vs time-weighted returns, Swensen has historically clipped material (100-500bps per annum) of additional returns via timely rebalancing. This wouldn't get picked up in a time-weighted composite return.
2. Final note, no endowment or foundation can make investment decisions based on something like how close the returns were to a 60/40. They still have to make a forward looking plan with all available information -- and importantly, how that informs their assumptions. So, to be able to invest that much capital in absolute return, for example, they are able to really reduce risk (defined in many ways) and most importantly, they have the best probability of maximizing expected returns given some level of maximum drawdown they are willing to undertake.

That was a lot of standing up for the endowment's out there who are able to invest like this (none really like Yale, but several more who absolutely add value via active management). I share the information only so those interested in reading those who write far less balanced articles have more insight in how this stuff can be misleading. There is simply more dispersion between these good endowments output and passive blends than reported numbers illustrate.

I do think you made a pretty fair case. I wish other writers in this space attempted to understand more about what's under the hood, and how/why annual report numbers may not do justice to that actual economic benefit they provide.

User avatar
Tyler9000
Posts: 376
Joined: Fri Aug 21, 2015 11:57 am

Re: We crush stock indexes, Yale claims

Post by Tyler9000 » Sat Mar 16, 2019 11:26 am

BJJ_GUY wrote:
Wed Mar 13, 2019 6:52 pm
This was a pretty interesting read, and unlike many who write about the topic, it seems you did the best with what you have to present fairly.

A few questions, and a couple comments to add some color:
1. When you are comparing drawdowns, are you simply linking annual returns? I'm assuming that's all you have, and if that's the case, your readers need to understand that those intra-period variance and drawdowns can be very different than what's stated. This is critically important due to the timing of the endowment's annual spend (which is pretty darn big for them).
Thanks for the feedback, BJJ_Guy. You make a lot of great points, and I completely agree that the temptation to over-simplify the complexities of managing a huge endowment can lead to a lot of hot takes that are ultimately unfair or shortsighted. My ultimate takeaway is not that either Buffett or Swensen are right or wrong, but that everyday investors should feel empowered by the fact that they can compete reasonably well with very smart investment managers simply by using intelligent asset allocation.

And regarding your question about drawdowns, your assumption about the data is correct and your warning about the absolute numbers is spot-on. More granular data is required if you want precise drawdown numbers in absolute terms, but in my experience annual drawdowns are still quite valuable for comparing the relative risk of different portfolios.

staythecourse
Posts: 6696
Joined: Mon Jan 03, 2011 9:40 am

Re: We crush stock indexes, Yale claims

Post by staythecourse » Sat Mar 16, 2019 12:38 pm

I believe Mr. Swensen is very honest that for the retail investor they have no shot at duplicating the Yale returns. So, unless that changes what is the point of arguing the passive vs. active argument?

BTW, I am so disappointed everyone compares their benchmark to the SP500. Unless you own 100% large cap stocks then you can't use the Sp500 as your benchmark. Endowments ESPECIALLY hold nothing similar to a sp500 portfolio. Mr. Swensen and anyone knowledgeable already knows that which means the article is a fluff piece.

Interesting, how strong active management is trying to discredit passive management. Maybe the greatest sign their is a large crack in their armor and real fear of their profession going forward?

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

yogesh
Posts: 286
Joined: Thu Oct 11, 2012 6:20 pm

Re: We crush stock indexes, Yale claims

Post by yogesh » Sat Mar 16, 2019 2:10 pm

Carthage College seems planning to go DIY with zero management fee for endowment.
After Abt retires in June, he may not be replaced. The board’s investment committee, made up of volunteers, would then take over the CIO duties. In an arrangement that would horrify Wall Street, they’d be managing money for free.
Emergency: FDIC | Taxable: VTMFX | Retirement: TR2040

inbox788
Posts: 5927
Joined: Thu Mar 15, 2012 5:24 pm

Re: We crush stock indexes, Yale claims

Post by inbox788 » Sat Mar 16, 2019 7:23 pm

yogesh wrote:
Sat Mar 16, 2019 2:10 pm
Carthage College seems planning to go DIY with zero management fee for endowment.
After Abt retires in June, he may not be replaced. The board’s investment committee, made up of volunteers, would then take over the CIO duties. In an arrangement that would horrify Wall Street, they’d be managing money for free.
That would be a terrific experiment. I hope they're able to establish a long term record so we can compare them to Yale and all the other college endowments. After fees (funds fees as well as advisor), I think they'll do quite well. Do Yale and other subtract out all the fees paid when they report the percentages? https://www.vox.com/2015/8/19/9178913/y ... ate-equity When they say "net of all fees", does that include the salary and expenses of Swensen and the rest of the investment managers at Yale? Or only 3rd party fees they paid out? May not be all that much for a multi-billion dollar endowment, but it might still make a couple basis points difference, and with smaller endowments, makes a much bigger percentage difference.

As far as beating Yale, I think much of their alpha comes from higher risk almost speculative investments that aren't accessible to passive investors. I don't think there's a low cost index fund for startup companies yet. There may be some startup ETFs one could invest in, but I think they're just active mutual funds wrapped up as ETFs with higher costs than index funds. Renaissance IPO is an interesting one though, with Yahoo Finance reporting 0.6 EF and tracking an actual post IPO index.

https://www.investopedia.com/articles/i ... artups.asp
https://etfdb.com/themes/artificial-intelligence-etfs/

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Tue Mar 19, 2019 12:32 pm

Tyler9000 wrote:
Sat Mar 16, 2019 11:26 am
BJJ_GUY wrote:
Wed Mar 13, 2019 6:52 pm
This was a pretty interesting read, and unlike many who write about the topic, it seems you did the best with what you have to present fairly.

A few questions, and a couple comments to add some color:
1. When you are comparing drawdowns, are you simply linking annual returns? I'm assuming that's all you have, and if that's the case, your readers need to understand that those intra-period variance and drawdowns can be very different than what's stated. This is critically important due to the timing of the endowment's annual spend (which is pretty darn big for them).
Thanks for the feedback, BJJ_Guy. You make a lot of great points, and I completely agree that the temptation to over-simplify the complexities of managing a huge endowment can lead to a lot of hot takes that are ultimately unfair or shortsighted. My ultimate takeaway is not that either Buffett or Swensen are right or wrong, but that everyday investors should feel empowered by the fact that they can compete reasonably well with very smart investment managers simply by using intelligent asset allocation.

And regarding your question about drawdowns, your assumption about the data is correct and your warning about the absolute numbers is spot-on. More granular data is required if you want precise drawdown numbers in absolute terms, but in my experience annual drawdowns are still quite valuable for comparing the relative risk of different portfolios.
I should have completed my though about using annual data, and looking at drawdowns as a gauge for some version of risk.

My point was more specific to how intra-period volatility - when combined with substantial capital outflows each year - can significantly impact the way in which we assess relative returns to passive blended benchmarks (or the policy portfolios).

We always calculate the TWR of a 70/30, or whatever, without regard for impact on the actual dollar returns impacted by timing and amount of flows. This isn't to be critical of your method, which is the best with what you have. It's only to say, that sometimes the portfolios that are 'working' don't necessarily compare as favorably in simple exercises like relative return analysis.

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Tue Mar 19, 2019 12:37 pm

staythecourse wrote:
Sat Mar 16, 2019 12:38 pm
I believe Mr. Swensen is very honest that for the retail investor they have no shot at duplicating the Yale returns. So, unless that changes what is the point of arguing the passive vs. active argument?

BTW, I am so disappointed everyone compares their benchmark to the SP500. Unless you own 100% large cap stocks then you can't use the Sp500 as your benchmark. Endowments ESPECIALLY hold nothing similar to a sp500 portfolio. Mr. Swensen and anyone knowledgeable already knows that which means the article is a fluff piece.

Interesting, how strong active management is trying to discredit passive management. Maybe the greatest sign their is a large crack in their armor and real fear of their profession going forward?

Good luck.
You realize Yale doesn't benchmark their public equity to the SPX, right? Even their Domestic Equity is against the Wilshire 5000 (not the 500). They use MSCI for their ex US equities.

I'm thinking maybe you are referring to articles being written about various institutions -- which tend to lazily use the SP500 as a simple equity proxy?? If so, I agree that it can be a bit misleading.

inbox788
Posts: 5927
Joined: Thu Mar 15, 2012 5:24 pm

Re: We crush stock indexes, Yale claims

Post by inbox788 » Tue Mar 19, 2019 1:19 pm

BJJ_GUY wrote:
Tue Mar 19, 2019 12:37 pm
staythecourse wrote:
Sat Mar 16, 2019 12:38 pm
I believe Mr. Swensen is very honest that for the retail investor they have no shot at duplicating the Yale returns. So, unless that changes what is the point of arguing the passive vs. active argument?

BTW, I am so disappointed everyone compares their benchmark to the SP500. Unless you own 100% large cap stocks then you can't use the Sp500 as your benchmark. Endowments ESPECIALLY hold nothing similar to a sp500 portfolio. Mr. Swensen and anyone knowledgeable already knows that which means the article is a fluff piece.

Interesting, how strong active management is trying to discredit passive management. Maybe the greatest sign their is a large crack in their armor and real fear of their profession going forward?

Good luck.
You realize Yale doesn't benchmark their public equity to the SPX, right? Even their Domestic Equity is against the Wilshire 5000 (not the 500). They use MSCI for their ex US equities.

I'm thinking maybe you are referring to articles being written about various institutions -- which tend to lazily use the SP500 as a simple equity proxy?? If so, I agree that it can be a bit misleading.
I don't see a problem using the SP500 as a benchmark for lack of a better single benchmark to compare all other investments to. Ultimately, we're interested in performance over the long term, and having a somewhat constant reference point is helpful. And it's a very readily available investment option that can be easily replicated, unlike some of the more esoteric and market timing strategies taken by the private equity hedge funds. Even the DJIA is adequate for some, and the correlation to the SP500 is fairly high. With the bulk of the SP500 and Wilshire5000 overlapping market cap wise, their performance isn't all that different. Take a look at the last 20 years and if you look at 10 years, it looks like even less difference.

http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Tue Mar 19, 2019 1:53 pm

inbox788 wrote:
Tue Mar 19, 2019 1:19 pm

I don't see a problem using the SP500 as a benchmark for lack of a better single benchmark to compare all other investments to. Ultimately, we're interested in performance over the long term, and having a somewhat constant reference point is helpful. And it's a very readily available investment option that can be easily replicated, unlike some of the more esoteric and market timing strategies taken by the private equity hedge funds. Even the DJIA is adequate for some, and the correlation to the SP500 is fairly high. With the bulk of the SP500 and Wilshire5000 overlapping market cap wise, their performance isn't all that different. Take a look at the last 20 years and if you look at 10 years, it looks like even less difference.

http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D
I think the argument would be to move in the opposite direction, a global index. The MSCI ACWI is more fitting than the S&P 500.

Because everyone likes to look back at just the last 10 years, there is a material upgrade in benchmark performance using the SP500 as US stocks have really outperformed non-US Equity during the trailing 10yrs. Think of all the articles that would be less sensational if they included the MSCI ex US as part of the benchmark proxy (or more simply just the ACWI). SP500 returned nearly 17% annualized over trailing 10 yrs (through Feb) compared to less than 10% for the MSCI ex US Index.

For this same backward-looking returns (and backward-minded extrapolations) there are a lot of retail investors thinking there is no reason for a global equity portfolio. Of course, if we look over rolling periods of returns we see that US and non US stocks tend to outperform one another at various points in history, with some periods being pretty long and pronounced. But, just as growth has been on a hell of a run versus value, there is no reason to believe anything other than mean reversion (based on price and value, not reversion for the sake of mean reversion, of course).

inbox788
Posts: 5927
Joined: Thu Mar 15, 2012 5:24 pm

Re: We crush stock indexes, Yale claims

Post by inbox788 » Tue Mar 19, 2019 2:49 pm

BJJ_GUY wrote:
Tue Mar 19, 2019 1:53 pm
I think the argument would be to move in the opposite direction, a global index. The MSCI ACWI is more fitting than the S&P 500.

Because everyone likes to look back at just the last 10 years, there is a material upgrade in benchmark performance using the SP500 as US stocks have really outperformed non-US Equity during the trailing 10yrs. Think of all the articles that would be less sensational if they included the MSCI ex US as part of the benchmark proxy (or more simply just the ACWI). SP500 returned nearly 17% annualized over trailing 10 yrs (through Feb) compared to less than 10% for the MSCI ex US Index.

For this same backward-looking returns (and backward-minded extrapolations) there are a lot of retail investors thinking there is no reason for a global equity portfolio. Of course, if we look over rolling periods of returns we see that US and non US stocks tend to outperform one another at various points in history, with some periods being pretty long and pronounced. But, just as growth has been on a hell of a run versus value, there is no reason to believe anything other than mean reversion (based on price and value, not reversion for the sake of mean reversion, of course).
I agree completely with the sentiment. Someday, ACWI may be where SP500 is to DJIA. Most media still focus on DJ, which for most practical purposes, is good enough. It may take some time for a world market index to build up a bigger following and have easily accessible long term data.

Just like for many uses, the metric system is simpler and superior, but we're still stuck on feet and inches, miles, pints and quarts, pounds and ounces,etc. At least investing is simpler than understanding tire sizes (mm, aspect ratio, speed rating, inches, load index, psi, etc.).

https://en.wikipedia.org/wiki/MSCI_World

http://quotes.morningstar.com/chart/etf ... 2%3A955%7D

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Tue Mar 19, 2019 2:59 pm

For long term returns, a good reference tool is the monthly update provided by AJO.

http://www.ajopartners.com/wp-content/u ... /19_02.pdf

Not sure if the above link will work. If it doesn't, use the one below and go to the 'print menu' section. Then select the current month (or whatever end point you want) for recent returns and the very long run returns for. wide number of indices.

http://www.ajopartners.com/print-menu/

staythecourse
Posts: 6696
Joined: Mon Jan 03, 2011 9:40 am

Re: We crush stock indexes, Yale claims

Post by staythecourse » Tue Mar 19, 2019 3:06 pm

BJJ_GUY wrote:
Tue Mar 19, 2019 12:37 pm
staythecourse wrote:
Sat Mar 16, 2019 12:38 pm
I believe Mr. Swensen is very honest that for the retail investor they have no shot at duplicating the Yale returns. So, unless that changes what is the point of arguing the passive vs. active argument?

BTW, I am so disappointed everyone compares their benchmark to the SP500. Unless you own 100% large cap stocks then you can't use the Sp500 as your benchmark. Endowments ESPECIALLY hold nothing similar to a sp500 portfolio. Mr. Swensen and anyone knowledgeable already knows that which means the article is a fluff piece.

Interesting, how strong active management is trying to discredit passive management. Maybe the greatest sign their is a large crack in their armor and real fear of their profession going forward?

Good luck.
You realize Yale doesn't benchmark their public equity to the SPX, right? Even their Domestic Equity is against the Wilshire 5000 (not the 500). They use MSCI for their ex US equities.

I'm thinking maybe you are referring to articles being written about various institutions -- which tend to lazily use the SP500 as a simple equity proxy?? If so, I agree that it can be a bit misleading.
Correct. I am getting annoyed of every analysis in articles using inappropriate benchmarks which no surprise always has to do with a 100% equity benchmark. As I mentioned in my post guys like Mr. Swensen are very keen on their portoflio does not follow a standard benchmark let alone a 100% SP500 benchmark.

The interesting thing is to hear from Mr. Swensen if knowing the above is true HOW do they measure their returns as good, bad, or average? That is another problem with active management. There is no way to know if you are doing well or still under performing your benchmark.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

User avatar
Taylor Larimore
Advisory Board
Posts: 27943
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Three Fund Portfolio beating Yale Portfolio

Post by Taylor Larimore » Tue Mar 19, 2019 3:10 pm

Bogleheads:

It is interesting to see that "Yale U's Unconventional" portfolio trails "Allan Roth's Second Grader Starter "Three Fund Portfolio" in the MarketWatch contest.

https://www.marketwatch.com/lazyportfolio

Best wishes
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

BJJ_GUY
Posts: 61
Joined: Wed Mar 13, 2019 7:45 am

Re: We crush stock indexes, Yale claims

Post by BJJ_GUY » Tue Mar 19, 2019 3:17 pm

staythecourse wrote:
Tue Mar 19, 2019 3:06 pm

Correct. I am getting annoyed of every analysis in articles using inappropriate benchmarks which no surprise always has to do with a 100% equity benchmark. As I mentioned in my post guys like Mr. Swensen are very keen on their portoflio does not follow a standard benchmark let alone a 100% SP500 benchmark.

The interesting thing is to hear from Mr. Swensen if knowing the above is true HOW do they measure their returns as good, bad, or average? That is another problem with active management. There is no way to know if you are doing well or still under performing your benchmark.

Good luck.
You lost me with the second paragraph. What do you think makes it so they can't compare manager or endowment level performance to their benchmark(s)? Everything is stated ex-ante, so it's not like they're gaming indices, policy benchmark, and/or the actual goal of the endowment (higher ed inflation + >spending rate).

I do agree completely that a global equity index is better than the SPX. But it's certainly not much use as a benchmark for endowments as a proxy for the underlying risk profile, at least not in periods shorter than looking over many sets of rolling 10 yr periods -- as an example.

User avatar
abuss368
Posts: 13414
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: Three Fund Portfolio beating Yale Portfolio

Post by abuss368 » Tue Mar 19, 2019 10:19 pm

Taylor Larimore wrote:
Tue Mar 19, 2019 3:10 pm
Bogleheads:

It is interesting to see that "Yale U's Unconventional" portfolio trails "Allan Roth's Second Grader Starter "Three Fund Portfolio" in the MarketWatch contest.

https://www.marketwatch.com/lazyportfolio

Best wishes
Taylor
Hi Taylor -

I have noticed this for a while now as I check the lazy portfolios every now and then. Interesting and more support for the Three Fund Portfolio. Shows investors more funds and complexity is not always better.

Best.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

Coltrane75
Posts: 26
Joined: Wed Feb 06, 2019 2:32 pm

Re: Three Fund Portfolio beating Yale Portfolio

Post by Coltrane75 » Wed Mar 20, 2019 9:02 am

abuss368 wrote:
Tue Mar 19, 2019 10:19 pm
Taylor Larimore wrote:
Tue Mar 19, 2019 3:10 pm
Bogleheads:

It is interesting to see that "Yale U's Unconventional" portfolio trails "Allan Roth's Second Grader Starter "Three Fund Portfolio" in the MarketWatch contest.

https://www.marketwatch.com/lazyportfolio

Best wishes
Taylor
Hi Taylor -

I have noticed this for a while now as I check the lazy portfolios every now and then. Interesting and more support for the Three Fund Portfolio. Shows investors more funds and complexity is not always better.

Best.
That's it folks, Yale beat passive indexing over one 20 year period, time for indexers to close it down. ;)

Post Reply