Wade Pfau looks at 4% WR, finds it unsafe

[Lsbcal]

I think intermediate gov bonds have negative real yield. The .3% for a broad portfolio doesn't seem off the mark to me.

Below is the 5 year TIPS rates. The historical average real rates for 5 year nominal Treasuries is about 2.3% (as per Swedroe) and TIPS will be below this by about the inflation rate plus a premium for taking inflation risk. We see that the last 10 years have seen a decline in the real rates and quite a dramatic decline over the last 3 years.

What could happen (among an infinite set of possibilities) is that we get back up to normal real rates in 3 to 5 years. Then we could even overshoot and go to fairly high real rates. The 1983-2000 period was a time of high real rates and we had a fairly good economy in those years. So there is a precedent.

None of us knows what the future will bring. Much of the SWR calculations (including Pfau's I think) assume a constant withdrawal rate. Many of us will adjust withdrawals to mitigate declining portfolios and as others have pointed out, that is not generally in those SWR models.

 

[Lsbcal]

I'm not even close to being a statistician (or mathematician) but can anyone comment on how Pfau's chart jibes with Firecalc's results?

I think Pfau is choosing selected stock/bond returns and then seeing how that plays out. FIRECalc is taking rolling periods with the actual historical results. Quite different methods.

At the same time, the bottom two curves flatten out around a 50% SA, suggesting a needless risk/reward. Counterintuitive?

Interesting question. I think one does not really know over a 20 or 30 year period exactly what their withdrawal rate (particular curve) will be. It might start out at 3% early on and then it could go up quite a bit (or down) depending on investment returns and life events. So you do not necessarily have to worry about being on those flat portions of one particular curve.

 

[Animorph]

About your 50% U.S. equities asset allocation: how many of those equities are shares of large multinational corporations with substantial earnings & assets from overseas business? I wonder what percentage of the S&P500 annual earnings comes from outside of the U.S.

And how many of my 50% foreign equities have substantial earnings and assets from U.S. business? It's pretty much just a convenient way to divide them up. However, they do behave differently and give me rebalancing opprtunities.

 

[audreyh1]

I'm anticipating, given the sluggish global economy and debt overhang, that interest rates don't normalize for quite a while - maybe not for another 10 years or more. It's in this environment that I wonder about portfolio survival. Currently, US companies are doing well, mortgages and REITs pay decent yields, so I'm not too worried.

 

[Lsbcal]

I'm anticipating, given the sluggish global economy and debt overhang, that interest rates don't normalize for quite a while - maybe not for another 10 years or more. It's in this environment that I wonder about portfolio survival. Currently, US companies are doing well, mortgages and REITs pay decent yields, so I'm not too worried.

I try to keep upbeat. Yes it's possible that real rates will stay low for many years. Hopefully the equity (US and international) returns will balance this out. That's why I've maintained a fairly high equity percentage in the AA.

We've noticed on a recent vacation how US workers, young and old, seem to be extremely well motivated to do a good job -- one of the upside benefits of a tight economy. One thing I think we can count on, the work world will continue to be an extremely competitive place. I'm hoping this translates into good equity returns.

 

[MichaelB]

He acknowledged this thread in his comments update. Hopefully he keeps reading. Wade, please feel free to join us and share your thoughts as your time permits.

I notice he lowers the expected return but keeps the same level of volatility for equities. Peter Bernstein said that lower annualized returns are the result of higher volatility, not lower annual returns. If that is the case (and I suspect it is, at least in part) the failure rate he shows for a 50/50 portfolio may be somewhat overstated. In other words, if disciplined rebalancers, allocation funds and mixed funds (Wellesley, Wellington) rebalance rigorously, those portfolios will survive longer.

 

[EvrClrx311]

The numbers in his assumption table appear inaccurate...

Stocks from 1926-2010
Arithmetic Mean: 8.70%
Geometric Mean: 6.62%

He then removes about 3.5% more off of those to account for inflation.

However, the numbers he started with (8.70%/6.62%) seem to already include inflation, since Compustat dataset of S&P500 from 1928-2003 lists:
Arithmetic Mean: 11.67%
Geometric Mean: 9.70%

missing 2004-2010 doesn't change things "that much"... it should be in the neighborhood of 11.3% and 9.5%

My guess is that Dr. Pfau counted inflation twice. There is no way to explain how a 3% withdrawal rate on a 90-10 (bond-stock) would fail over 30% of the time over a 30 year period using the median returns of the market over the last 80 years. Inadvertently setting inflation to double what it actually is... would accomplish that.

Another supporting factor is that his table shows a higher failure rate with bonds taking over the majority of holdings. That kind of strong trend would only happen if bonds significantly trailed the inflation used...

Only other explanation is that he may be excluding dividends being reinvested for the stock portion... which historically have accounted for a return in the neighborhood of inflation over time.

 

[EvrClrx311]

Found it... the annualized return of the S&P500 from 1926-2010 was 9.87%

Even if you cherry picked the absolute bottom of the market in March 2009 (when it fell 50%)... essentially the worst 83 year return we'll ever see for the next 50+ years, the annualized market return from 1926-March 2009 was still 6.98%

So I'm not sure where he got the 6.62% geometric mean number from... it must be missing something (either dividends or inflation) already.

 

[unclemick]

Agile, mobile and hostile. After 18 years, I can't even spell engineer any more so I avoid taking numbers real serious. My unified theory of chickenheartedness requires that I continue to charge into the future with one finger in my belly button and one eye on the rear view mirror. Having roared past 55 and 62 in ER - I now have a floor, SS and non-cola pension, aka 100 to 30% of expenses depending on my navel quiver from last years portfolio. 4% centerpoint, actual takeout 2 to 6% range based on last years performance.

Old habits mean I run FireCalc, ORP several times during the year and avidly read threads like this one.

However I take to heart Dory33's original admonition - to be careful when measuring with a micrometer and cutting with an axe.

Handgrenade and horseshoe wise I run calculators and look at studies to feel warm and smarmy about being in the ballpark.

heh heh heh - put me in the floor and agressively varying expenses year to year school - cause I've given up (many decades ago) trying to outsmart Mr Market with portfolio shifts.

 

[Lsbcal]

Found it... the annualized return of the S&P500 from 1926-2010 was 9.87%

Even if you cherry picked the absolute bottom of the market in March 2009 (when it fell 50%)... essentially the worst 83 year return we'll ever see for the next 50+ years, the annualized market return from 1926-March 2009 was still 6.98%

So I'm not sure where he got the 6.62% geometric mean number from... it must be missing something (either dividends or inflation) already.

Interesting observation.

My data for the SP500 with dividends from 1951 to present is CAGR = 10.5%. For this period the inflation rate was 3.6% so roughly the real SP500 + dividends return was CAGR = 6.9%.

 

[Lsbcal]

...(snip)...
Old habits mean I run FireCalc, ORP several times during the year and avidly read threads like this one.

However I take to heart Dory33's original admonition - to be careful when measuring with a micrometer and cutting with an axe.

Handgrenade and horseshoe wise I run calculators and look at studies to feel warm and smarmy about being in the ballpark.

heh heh heh - put me in the floor and agressively varying expenses year to year school - cause I've given up (many decades ago) trying to outsmart Mr Market with portfolio shifts.

Wise comments. I like the Dory33 quote too. Perhaps you should write a blog on this stuff but that might be work .

 

[REWahoo]

Hey guys, I know it's been a while since Dory36 moved on to find greener pastures, but please don't use that axe on his username.

 

[Nords]

Much of the SWR calculations (including Pfau's I think) assume a constant withdrawal rate.

Yep.

I'm anticipating, given the sluggish global economy and debt overhang, that interest rates don't normalize for quite a while - maybe not for another 10 years or more.

I think an analogy is after WWII when the govt did everything they could to keep interest rates as low as possible for as long as possible in order to get rid of all the War Bond debt.

The federal govt has plenty of incentive to keep rates low, even if inflation kicks up to 4%. I bet they've been debating this scenario since March 2009...

 

[Gumby]

Yep.


I think an analogy is after WWII when the govt did everything they could to keep interest rates as low as possible for as long as possible in order to get rid of all the War Bond debt.

The federal govt has plenty of incentive to keep rates low, even if inflation kicks up to 4%. I bet they've been debating this scenario since March 2009...

An interesting paper on this very topic from the National Bureau of Economic Research.

http://www.imf.org/external/np/seminars/eng/2011/res2/pdf/crbs.pdf

 

[Lsbcal]

An interesting paper on this very topic from the National Bureau of Economic Research.

http://www.imf.org/external/np/seminars/eng/2011/res2/pdf/crbs.pdf

Reading the abstract from that paper, they mention "financial repression". This is a term the Pimco (Gross, et. al.) has used a lot recently in regard to the current bond scenario.

If this continues for some years there are going to be a lot of unhappy fixed income investors. So far many of those have benefited somewhat from the falling rate environment but that appears to have played out.

From my SWR models, a particular trying period for a stock/bond portfolio was 1946 to 1949 -- just after WW2. This was true whether your portfolio was 60/40 or 40/60.

 

[Nords]

An interesting paper on this very topic from the National Bureau of Economic Research.
http://www.imf.org/external/np/seminars/eng/2011/res2/pdf/crbs.pdf

Historically, periods of high indebtedness have been associated with a rising incidence of default or restructuring of public and private debts. A subtle type of debt restructuring takes the form of “financial repression.” Financial repression includes directed lending to government by captive domestic audiences (such as pension funds), explicit or implicit caps on interest rates, regulation of cross-border capital movements, and (generally) a tighter connection between government and banks. In the heavily regulated financial markets of the Bretton Woods system, several restrictions facilitated a sharp and rapid reduction in public debt/GDP ratios from the late 1940s to the 1970s. Low nominal interest rates help reduce debt servicing costs while a high incidence of negative real interest rates liquidates or erodes the real value of government debt. Thus, financial repression is most successful in liquidating debts when accompanied by a steady dose of inflation. Inflation need not take market participants entirely by surprise and, in effect, it need not be very high (by historic standards). For the advanced economies in our sample, real interest rates were negative roughly ½ of the time during 1945-1980. For the United States and the United Kingdom our estimates of the annual liquidation of debt via negative real interest rates amounted on average from 3 to 4 percent of GDP a year. For Australia and Italy, which recorded higher inflation rates, the liquidation effect was larger (around 5 percent per annum). We describe some of the regulatory measures and policy actions that characterized the heyday of the financial repression era.

Thanks, that's exactly what I read from some financial blog.

Like they say, "hope for growth" is not a plan...

 

[Lsbcal]

Thanks, that's exactly what I read from some financial blog.

Like they say, "hope for growth" is not a plan...

Exactly. Now we see Francois Hollande newly elected in France touting a "growth" agenda. The Germans say they might agree to some growth measures but do not want to renegotiate the austerity agreements.

We all want growth of our economies but there are many agendas for achieving that state. In the US Presidential race we'll see plenty of growth agenda talk. Pick your poison.

 

[unclemick]

Hey guys, I know it's been a while since Dory36 moved on to find greener pastures, but please don't use that axe on his username.

Oops! Senior moment - Dory 36's - 33% That's my story convinced me in ancient times to join this form.

heh heh heh - what's a little mental gear separation among friends. .

 

[Midpack]

I think he's still working on it, although he seems to be experimenting with some simulations.

I'm not expecting a response, but FWIW I sent Wade Pfau an email yesterday asking him about floor income as a plan A (his plan evidently) versus plan B (my plan until I threaten my annuitization hurdle, and I linked the fpanet article as a reference). We'll see...

UPDATE: I did get a response! http://www.early-retirement.org/forums/f28/annuity-as-plan-a-versus-plan-b-61602.html

 

[MichaelB]

The comments section of his blog post show he is reflecting on some of the observations here

At some point it is worth discussing further, as I think a lot of confusion arises since it is not always clear when someone is talking about real or nominal returns, and when someone is talking about arithmetic or geometric/compounded returns.

Here I am showing only real returns, but you could see the nominal returns by adding the inflation numbers to the real returns.

I didn't double count inflation, the stock return falls so much in Table 3 because it is tied to the bond return through the equity premium, and I lowered the bond return to reflect current conditions and lowered the equity premium to reflect the international average.

I'm running this a few more times now and I do see that the failure rates are still bouncing around a bit from simulation to simulation. The minimum failure rate for 3% is coming out in the range 12-16%, and 4% has failures in the neighborhood of 35-39%. It was merely a coincidence, but the version of the figure I posted does seem to be leaning toward the high failure rate side. Is that what you are finding? It is always good to have replications.

This is encouraging. Too much of the academic work in this area doesn't seem to apply to our real world experience and he is attempting to make this relevant to us.

 

[EvrClrx311]

Interesting observation.

My data for the SP500 with dividends from 1951 to present is CAGR = 10.5%. For this period the inflation rate was 3.6% so roughly the real SP500 + dividends return was CAGR = 6.9%.

fwiw, the number I quoted (6.98 being the worst) was without inflation factored out yet (converted to real SP500 + dividends its CAGR was about 3.38%). So your 6.9% number is approximately 3.5% higher than the absolute market minimum return I was alluding to. The one that ended on the day the market hit something like 6500 (rock bottom).

A 7% real market return for stocks is pretty consistent historically when looking at very long term returns 30+ years... and not purposefully selecting a start or end point at a top or bottom. Realistically... no one is going to sell their entire portfolio on a day like March 9th 2009. As you can see now, it only took 2-3 years for the market to double and bounce back from that. Also, no one is going to invest all of their cash into stocks on the absolute height... the gradual in and out helps to mute these extreme cases and show that in reality you'll absolutely never see a 30 year real equity return anywhere close to as low as 3.6%

Using a number as low as that is incredibly pessimistic (beyond reasonable) compared to historical standards... to be ultra conservative he should maybe try 5% instead, which would show that the golden 4% withdrawal is most certainly is "safe"

 

[samclem]

A 7% real market return for stocks is pretty consistent historically when looking at very long term returns 30+ years...

I take it you are talking about US returns (among the world's most productive economies and one that arguably enjoyed advantages it no longer does) over the last 100 years (the most productive such period in history).
What appears to be a "big sample" to some people may properly be viewed as cherry-picking given other perspectives.

 

[EvrClrx311]

I take it you are talking about US returns (among the world's most productive economies and one that arguably enjoyed advantages it no longer does) over the last 100 years (the most productive such period in history).
What appears to be a "big sample" to some people may properly be viewed as cherry-picking given other perspectives.

very similar things were said in 1873, 1896, 1913, 1929, 1954, 1973, 2000

"this time its different..." is nothing new. It's human nature to think what we're experiencing today won't fit the models of yesterday... that somehow we'll never get back to the prosperity we saw before. It's the very thing that by nature makes humans horrible investors.

I challenge you to bookmark this comment and come back to look at it in 20 years (10 years might even be enough to show it). I can all but guarantee you that we'll see 5%+ real market returns (8.5%+ actual) between now and then for the S&P500.

If somehow you are correct that the next 20 years fall outside the rules the last 250+ have followed, then I'll call you a genius... or lucky. Either way you can have a good laugh at my expense.

 

[Lsbcal]

...(snip)...
I can all but guarantee you that we'll see 5%+ real market returns (8.5%+ actual) between now and then for the S&P500.
...

Shame on you for being so positive.
Despite your hopeless optimism, I applaud the sentiment.
My portfolio is nodding in agreement too.

 

[EvrClrx311]

Shame on you for being so positive.
Despite your hopeless optimism, I applaud the sentiment.
My portfolio is nodding in agreement too.

I've always had an extreme interest in following statistics, investing, human nature and psychology... I often find myself at odds with the notion of "common sense":

In 1998 (late fall) ... as a term project I argued that a recession was statistically a certainty, coming soon. "Recession: All but guaranteed" (lol, I used those exact words as the title of my paper) ... my teacher gave me an B+ overall with a C for message content dragging it down. He thought I was so off-base with the 'prediction.' I don't think he bothered to look at any of the supporting evidence I provided. I wrote him a couple years later and he said he didn't remember my paper or the grade he had given me...

In 2005, I purchased my first house and had a lot of engaging discussions with my realtor trying to understand how it was housing could constantly beat inflation ("the fact that it has risen so much recently... means won't it most likely have horrible return for the next 30 years?" I asked)... she gave me a blank stare almost as if to say, "did you seriously just ask me that? Look at how hot housing is right now!"

In February 2009... I took all the cash I had on hand, as well as a fairly large inheritance, and invested it into stocks/equities. I advised everyone I knew that this was a once in a generation (possibly lifetime) situation to invest in the stock market when it would bring huge returns within 4 years (the market never takes longer than that to recovery from an extreme drop like what we saw at the time)


Looking back... all of those seem reasonable and not all that incredible to predict. I can tell you that at the time of each I was viewed as a looney...