University Endowment Investment and Disbursement

[haha]

Big Squeeze on Ivy League Endowments - Barrons.com

Barron's has an intersting article on university endowments this weekend. It might take a sub to see it. If so, maybe someone with more web skills than I can find another free location.

What caught my eye is that these endowments are intended to be perpetual, yet they must disburse an average of 5% annually. So in a sense they are in the position of a young ER who chooses a 5% withdrawal rate.

Then are in the active withdrawal phase, but they do not intend to be in drawdown. The big private ones- Harvard, Yale, Princeton, Stanford-are are quite diversified, but employ almost no fixed income (other than perhaps aggressive hedge fund fixed income strategies).

If they were using an approach similar to the lifecycle espoused by FAs etc, they would need at least 50% bonds. The article points out that theses funds have had a tough time the last few years. [Yale's] "Swensen acknowledged that "diversification isn't going to help you in the midst of a financial crisis, or at least the type of diversification that you see in institutional portfolios like Yale's." He added, "I'm not sure that the crisis has caused us to conclude that we would do things differently, but it certainly highlighted the importance of liquidity."

Don't know what all this might mean for the individual retiree, but it is perhaps food for thought.

Ha

 

[LOL!]

Maybe you didn't go to a university with a big endowment? They are NOT like an early retiree with no income. They are CONTINUALLY trolling for more donations which can be substantial, so they have plenty of income going into the endowment.

 

[W2R]

Yes, it's as though they were still working while drawing down. In fact, their income probably increases year after year.

 

[FUEGO]

And I think the analogy to a young very early retiree would have to be someone who decided to take a 5% withdrawal of CURRENT portfolio value instead of the typical 4% SWR adjusted annually for CPI inflation. In other words, the endowments only have to pay out 5% of whatever they have left (IIRC from my class on perpetual trusts). If they are down to their last buck, they only have to pay out $0.05 that year.

A withdrawal strategy that is based on taking a percentage of portfolio value each year will never technically "fail" in the sense that you can always take 5% of something, even if it won't buy you much. Obviously it fails as a practical matter since one cannot readily live on pennies a year.

The low allocation to bonds definitely makes sense (IMHO) for a very young retiree who is essentially trying to establish a perpetual portfolio. Historically, short to intermediate term treasuries have not returned more than 1-2 percent over inflation (right?), so real returns are rather low. Going up the risk scale and maturity scale increases returns a little, but doesn't generally compensate for added volatility from a risk adjusted return perspective.

Most investments in these endowment portfolios seem to be of the high expected return but volatile types (equities, hedge, private equity), while also seeking low levels of correlation between asset classes and maintaining a good hedge against inflation.

 

[Independent]

I think this is interesting. Retirees want stable incomes, but they often invest in very volatile assets to support that income. The do so because they think that "over the long run" the volatile assets will have higher yields. Many of the posts on this forum discuss ways to reconcile the conflict. Universities have the same problem.

But, it seems to me there was a thread on this back in May http://www.early-retirement.org/forums/f28/harvard-rejects-withdrawal-premise-of-firecalc-44324.html

 

[FUEGO]

I think this is interesting. Retirees want stable incomes, but they often invest in very volatile assets to support that income. The do so because they think that "over the long run" the volatile assets will have higher yields. Many of the posts on this forum discuss ways to reconcile the conflict. Universities have the same problem.

But, it seems to me there was a thread on this back in May http://www.early-retirement.org/forums/f28/harvard-rejects-withdrawal-premise-of-firecalc-44324.html

Yeah, a thread also by Haha, too.

It is an interesting trade-off between potential for better long term return and very short term concerns of year to year stability in distributions.

A number of solutions to this problem have cropped up over the years. There's ESRBob's (Bob Clyatt) 95% rule (based on previous research I believe). There's the Ray Lucia buckets of money that allow limited stability of income (7 years??) and fairly stable income 7 more years. Then there's the classic norwegian widow "live on dividends" and maybe interest payments.

One idea I have bounced back and forth on here over the years is what I call the "hybrid" withdrawal strategy that involves a portion of the withdrawal (call it 'X') being based on a fixed percentage of initial portfolio value and the remainder of the withdrawal (call it 'Y') based on a percentage of the portfolio value each year. Let's just say X=2% fixed withdrawal and Y=2% variable withdrawal. You are guaranteed to have the 2% inflation adjusted withdrawal every year, plus 2% of whatever your porfolio is worth any given year. In bad years you tighten your belt, get a part time income, ebay some junk, cut back on vacations, delay maintenance items, delay capital purchases and capital improvements, etc. In good years, you can take much nicer vacations, catch up on capital improvements, capital purchases (ex: cars), maintenance, etc.

This hybrid withdrawal plan is sort of a compromise between long term growth and current stability of withdrawal amounts. It allows you to take advantage of higher spending that is made possible by long term real growth of the portfolio value (something not strictly allowable per the 4% of initial portfolio value plus annual CPI adjustments that is the "4% rule"). At the same time the inclusion of the fixed component of 2% of initial portfolio value dampens the year to year volatility of the withdrawals.

For example, imagine someone following the hybrid withdrawal strategy with 2% fixed and 2% variable withdrawals. Assume an 80%/20% stock/bond portfolio overall. The stock side lost 40% in 2008 and the bond side gained 2%. Overall the 80/20 portfolio lost 32% of its value last year.

The 2% fixed component continues to allow 2% withdrawals this year. The 2% variable component is based on the new portfolio value that is conveniently 32% smaller. So instead of a 2% variable withdrawal (based on initial portfolio value), you only get 2% of what is left, or 1.37% based on initial portfolio value. Overall, your 4% withdrawal rate just dropped to 3.37%, a 16% drop in spending power. That sucks. But not the end of the world unless you ER'd on a barebones budget in the first place. Your hypothetical $50,000 annual withdrawal just turned into a $42,100 annual withdrawal, so you cut $7,900 out of your budget (or get a part time job). And this is following one of the worst economies and market declines in recent history.

Obviously you could still have a failure. Prolonged sideways or downwards markets would make your real annual withdrawal wither away (along with your portfolio).

 

[haha]

Maybe you didn't go to a university with a big endowment? They are NOT like an early retiree with no income. They are CONTINUALLY trolling for more donations which can be substantial, so they have plenty of income going into the endowment.

Not news to me. However, if you look at what Harvard gets each year compared to what they spend, they are still drawing 5% from the endowments. These schools would be nowhere close to self supporting from tuition and annual alumni giving.

Fuego and Independent- true, they are using 5% of the current value, rather than some "adjusted starting value". Very different. I just thought it was interesting, not definitive or exactly comparable to a retirees situation.

Ha

 

[audreyh1]

I don't see any parallel between the retiree who is trying to achieve some stable withdrawal each year which tracks with inflation without running out of money over a fixed time period (which is where we get the Trinity study approach) and what these perpetual endowment funds try to achieve.

Personally, I've chosen the "endowment" percentage-recalculated-each-year approach which means I have to be willing to live with wild swings in the amount withdrawn each year. But that doesn't bother me as it immediately adjusts for bad market conditions as well as providing a bigger windfall under good market conditions. At the gut level this approach just "feels" more reasonable (you can't in fact run out of money), especially since I am still looking at 40+ years.

From what I have read, in the 2000s pension funds and endowments started using increasingly exotic investments such as hedge funds and private equity. I think a lot of them really got burned badly with the recent financial crisis. And certainly, if like Harvard your portfolio is pretty much 100% equities, even though allocated among different types of equities, well you aren't going to see any cushion from diversification during a financial crisis.

I think these funds have just gotten "too smart" for their own good and abandoned sensible diversification principles. I too am surprised that bonds don't figure as a large percentage of the allocation. I guess they think that the hedge funds will provide sufficient "hedging" against the rest of the portfolio? But even many hedge funds over the past decade abandoned the traditional hedging approach.

Audrey

 

[Independent]

Yeah, a thread also by Haha, too.

It is an interesting trade-off between potential for better long term return and very short term concerns of year to year stability in distributions.

A number of solutions to this problem have cropped up over the years. There's ESRBob's (Bob Clyatt) 95% rule (based on previous research I believe). There's the Ray Lucia buckets of money that allow limited stability of income (7 years??) and fairly stable income 7 more years. Then there's the classic norwegian widow "live on dividends" and maybe interest payments.

One idea I have bounced back and forth on here over the years is what I call the "hybrid" withdrawal strategy that involves a portion of the withdrawal (call it 'X') being based on a fixed percentage of initial portfolio value and the remainder of the withdrawal (call it 'Y') based on a percentage of the portfolio value each year. Let's just say X=2% fixed withdrawal and Y=2% variable withdrawal. You are guaranteed to have the 2% inflation adjusted withdrawal every year, plus 2% of whatever your porfolio is worth any given year. In bad years you tighten your belt, get a part time income, ebay some junk, cut back on vacations, delay maintenance items, delay capital purchases and capital improvements, etc. In good years, you can take much nicer vacations, catch up on capital improvements, capital purchases (ex: cars), maintenance, etc.

This hybrid withdrawal plan is sort of a compromise between long term growth and current stability of withdrawal amounts. It allows you to take advantage of higher spending that is made possible by long term real growth of the portfolio value (something not strictly allowable per the 4% of initial portfolio value plus annual CPI adjustments that is the "4% rule"). At the same time the inclusion of the fixed component of 2% of initial portfolio value dampens the year to year volatility of the withdrawals.

For example, imagine someone following the hybrid withdrawal strategy with 2% fixed and 2% variable withdrawals. Assume an 80%/20% stock/bond portfolio overall. The stock side lost 40% in 2008 and the bond side gained 2%. Overall the 80/20 portfolio lost 32% of its value last year.

The 2% fixed component continues to allow 2% withdrawals this year. The 2% variable component is based on the new portfolio value that is conveniently 32% smaller. So instead of a 2% variable withdrawal (based on initial portfolio value), you only get 2% of what is left, or 1.37% based on initial portfolio value. Overall, your 4% withdrawal rate just dropped to 3.37%, a 16% drop in spending power. That sucks. But not the end of the world unless you ER'd on a barebones budget in the first place. Your hypothetical $50,000 annual withdrawal just turned into a $42,100 annual withdrawal, so you cut $7,900 out of your budget (or get a part time job). And this is following one of the worst economies and market declines in recent history.

Obviously you could still have a failure. Prolonged sideways or downwards markets would make your real annual withdrawal wither away (along with your portfolio).

One of the best insights I've heard on retirement spending is from a co-worker who said that you have to be able to distinguish between your wants and your needs. You should plan to cover your needs almost all the time, but you can accept volatility in funding your wants. I think that's what nearly everyone on this board is doing, but they have different methods (including "I'll cross that bridge when I get to it").

I thought your approach was interesting enough to try to do some numbers. I'll assume I have $1 million saved. Using the 4% SWR strategy, I withdraw $40k per year, good times and bad. That says to me that I "need" to have the full $40k. If I can flex down, maybe as low as $20k, I should be able to spend more on average.

I've got a little Monte Carlo system that works like FIRECALC, but that I can modify as I like. I set it up with a mean and std dev of investments (6.8% and 12.4%) that produced exactly 50 failures in 1000 runs using the 4% SWR rule.

Then I set up what you described, taking an inflated $20k + 2.0% of the current fund every year. The results where about what you'd expect (you may have done this already). For example, in the 5th year, two-thirds of the paths gave me more than $40k withdrawals, with an average of $45k. The other third were below $40k, with an average of $37k. If I live 30 years, the last year has 80% of the paths above $40k, averaging $79k, with the other 20% averaging $32k. I had only one failure in 1000 paths, compared to 50 failures with the 4% SWR.

With those results, it seems reasonable to try higher initial withdrawals. So I tried $25K + 2.5%. The resulting 5th year numbers were 98% averaging $52k and 2% averaging $39k. By the 30th year, it's 79% averaging $79k, 18% averaging $34k, and 3% failures.

So the willingness to flex down to $25k if times are pretty bad is enough to move the early payouts from a fixed $40k to a variable amount that averages over $50k. It certainly looks like something people should think about.

 

[FUEGO]

Then I set up what you described, taking an inflated $20k + 2.0% of the current fund every year. The results where about what you'd expect (you may have done this already). For example, in the 5th year, two-thirds of the paths gave me more than $40k withdrawals, with an average of $45k. The other third were below $40k, with an average of $37k. If I live 30 years, the last year has 80% of the paths above $40k, averaging $79k, with the other 20% averaging $32k. I had only one failure in 1000 paths, compared to 50 failures with the 4% SWR.

With those results, it seems reasonable to try higher initial withdrawals. So I tried $25K + 2.5%. The resulting 5th year numbers were 98% averaging $52k and 2% averaging $39k. By the 30th year, it's 79% averaging $79k, 18% averaging $34k, and 3% failures.

So the willingness to flex down to $25k if times are pretty bad is enough to move the early payouts from a fixed $40k to a variable amount that averages over $50k. It certainly looks like something people should think about.

This is basically what I determined too. You have a pretty good chance of better long term withdrawals and less risk of technical failure. Like you say, you have to be willing to live with the $25k payout instead of the $40k payout if times get really tough. But that probably represents a reasonable spread between needs (=$25k) and wants (+$15k). And in average or good times, you can live at a higher standard of living. I think this "hybrid" withdrawal strategy with one part fixed and one part variable more closely represents the spending patterns of real ER's. That is, people spend less when their portfolios go down (at least on discretionary expenses), and spend more if portfolio values go up.

I think if you examine the time series of portfolio values, you will see that the critical point is the first 10-15 years. After that initial period of the portfolio values holding up okay, you have a much better chance of long term survival (because the principal value has grown sufficiently to provide a wide margin of safety in most cases). And for ER's, we often have "go back to work" in the back pocket as plan C or plan D if plan A and B don't work out (at least for the first 10 years or so - YMMV of course).

 

[Ed_The_Gypsy]

IMHO, LOL! nailed it.

 

[free4now]

I've never really bought the idea of "flexing" down from a $40k to a $25k lifestyle as some would advocate. It seems most of the "flex down" proponents haven't really thought through the frictional costs of selling a home, moving to a new area, buying a new one, and furnishing the new one. That can easily eat up a year's worth of spending, which most who are facing declining assets are leary to do.

Unless you do a lot of travelling or other luxurious discretionary spending, I don't think most people can flex down much. Going back to part or full time work is much more feasible.

 

[Independent]

I've never really bought the idea of "flexing" down from a $40k to a $25k lifestyle as some would advocate. It seems most of the "flex down" proponents haven't really thought through the frictional costs of selling a home, moving to a new area, buying a new one, and furnishing the new one. That can easily eat up a year's worth of spending, which most who are facing declining assets are leary to do.

Unless you do a lot of travelling or other luxurious discretionary spending, I don't think most people can flex down much. Going back to part or full time work is much more feasible.

This is why it's important to separate wants and needs when you are planning. I'll agree that someone who doesn't have any fat in the budget has to take the conservative route. Someone who can see where the spending could be cut can be more aggressive in the early years. I think you're saying that some people think they can cut, but haven't really looked at what they'd have to give up to make a big dent - that's probably true, and it's a problem for them.

In the $1 million case that I've thought about, I was assuming the couple had $30k of Social Security as a base. So for that family it's a question of a stable $70k total, or a variable amount that starts at $80k. In the second case, the first $10k of flex simply gets them down to the $70k.

You're correct about travel. That's the area that I automatically drop into the "wants" category. Some people don't have any of that in the budget.

 

[audreyh1]

I think it was cute fuzzy bunny or the trout dude or someone who preached having a retirement nest egg that could fund 2x the absolute bare minimum needs. That way under normal circumstances you can have plenty for "wants", but in hard times you can severely cut back without undue hardship.

Audrey

 

[LOL!]

So this means that the frugalistas have to change their lifestyles and double their spending by adding an equal amount of frivolous expenses? (Unless we are in hard times now?)

 

[audreyh1]

So this means that the frugalistas have to change their lifestyles and double their spending by adding an equal amount of frivolous expenses? (Unless we are in hard times now?)

No, if you are a bare bones frugalista saving for retirement you might just want to save enough so that you can double up your spending after retirement if the mood strikes you.

Audrey

 

[growing_older]

Or you may want to consider adding to your wants as you age. For instance, as much as I enjoy gardening, I expect I will eventually want to hire someone to do yardwork when I am old enough not to enjoy heavy lifting. Having the means to do that isn't important to an ER, but it may be important later in retirement.

 

[audreyh1]

Travel is another big potential expense in retirement. With limited vacations while working, there is not much opportunity to travel extensively (and work travel doesn't count 'cause you don't really get to enjoy the benefits of travel, and certainly not leisurely travel).

Audrey

 

[FUEGO]

I've never really bought the idea of "flexing" down from a $40k to a $25k lifestyle as some would advocate. It seems most of the "flex down" proponents haven't really thought through the frictional costs of selling a home, moving to a new area, buying a new one, and furnishing the new one. That can easily eat up a year's worth of spending, which most who are facing declining assets are leary to do.

Unless you do a lot of travelling or other luxurious discretionary spending, I don't think most people can flex down much. Going back to part or full time work is much more feasible.

Sure, if the $40k is a barebones budget and you retired with just enough to cover a barebones budget, then you have little room to "flex down". But if $25k reflects just your needs, you can maybe get by for a while at that level.

The $15k deficit can be made up for by reducing expenses or taking on part time work. Say you cut out $7500 in expenses (reduced travel, less gifts, deferred maintenance on house/cars, no remodeling, no new cars, less or cheaper dining out) and you pick up a part time job that nets you another $7500 a year. You are back to where you started at $40k/yr. Even at minimum wage, that $7500 a year is only 1000 hrs of labor. Split between two people, that is 1/4 time (10 hrs a week). Make double min wage, and you are down to 20 hrs/month.

Whether you go with some sort of fixed or variable or hybrid withdrawal strategy, there is going to be a lot of pressure to minimize spending when your portfolio takes a bath (think back to March 2009!). You can either set up your withdrawal strategy to explicitly account for this (by making withdrawals somehow variable), or you can implicitly acknowledge that you may need to reevaluate your withdrawal levels if the market tanks. Either way, you are left with the ability to spend more in good times and the fear to spend less in bad times.

 

[jdw_fire]

I think it was cute fuzzy bunny or the trout dude or someone who preached having a retirement nest egg that could fund 2x the absolute bare minimum needs. That way under normal circumstances you can have plenty for "wants", but in hard times you can severely cut back without undue hardship.

Audrey

it was cut throat.


how bout the buy a COLAd SPIA to cover "needs" and use the rest of your portfolio for wants strategy?

 

[Independent]

how bout the buy a COLAd SPIA to cover "needs" and use the rest of your portfolio for wants strategy?

To me, this should be the "base strategy" for most people. The neat thing is that once you've covered your needs, you can spend the rest of your money in whatever pattern you like -- spend half of it on an around-the-world cruise in the first year if that strikes your fancy. Certainly, you don't need to go with the very conservative 4% withdrawals on the rest, 8% makes more sense.

By "base strategy" I mean that you do the calculation even if you don't think this is where you're going to end up. It helps clarify your goals and it provides realistic numbers. One problem is that private COLA'd SPIAs aren't very common today, but anybody can defer SS and get the same result.

Most people don't do this. I'd guess the big problems are that people don't like to "lose control" of their money, and they have trouble identifying "wants" that they are willing to put at risk.

 

[audreyh1]

it was cut throat.

Thanks! Yeah, for the life of me I couldn't remember his forum name. I knew it was some type of trout!

Audrey

 

[Gone4Good]

So this means that the frugalistas have to change their lifestyles and double their spending by adding an equal amount of frivolous expenses? (Unless we are in hard times now?)

Not at all. But I think it's important for people to realize that a 4% SWR that just covers basic needs is a heck of a lot more risky than a 4% SWR with plenty of fat built in.

 

[Free To Canoe]

Excerpts from the article:

University Endowments, "The Big Squeeze" - Wall Street Examiner Forums

Free to canoe