If you missed the 10/31/17 Reddit “Ask Me Anything” interview with Dr. Wade Pfau, you really should read it. But if you just want the highlights, the following Q&As are the ones I myself found most valuable.
Dr. Pfau is a major post-William Bengen contributor to the financial independence/early retirement movement, especially with respect to safe withdrawal rates. To quote his bio:
Wade D. Pfau, Ph.D., CFA, is a Professor of Retirement Income in the Ph.D. program for Financial and Retirement Planning at The American College in Bryn Mawr, PA. He also serves as a Principal and Director for McLean Asset Management. He holds a doctorate in economics from Princeton University and publishes frequently in a wide variety of academic and practitioner research journals on topics related to retirement income. He hosts the Retirement Researcher website, and is a contributor to Forbes, Advisor Perspectives, Journal of Financial Planning, and an Expert Panelist for the Wall Street Journal. He is the author of the books, How Much Can I Spend in Retirement? A Guide to Investment-Based Retirement Income Strategies, and Reverse Mortgages: How to Use Reverse Mortgages to Secure Your Retirement.
For the sake of convenience I’ve attempted to group and parse these Q&As without detracting from their meaning. I may have screwed that up, of course. Call me out in the comments if you think that’s the case.
Audience questions are bolded; Pfau’s answers are plain text.
What advice do you have for young adults when it comes to planning out for retirement?
-Focus on Financial independence. You may or may not want to retire when the time comes, but with financial independence you will have the freedom to do what you want.
-Make sure you have a plan for how you will spend your time. Wanting to be retired is more important then just not wanting to work. You’ve got to do something.
-Nothing beats the old: save early and often. The needed savings rate doubles for each 10 years you wait to get started.
What are your honest thoughts on our basic FIRE strategy? Is there a way we can improve upon it?
It seems like a lot of the FIRE community is what I call “probability-based,” i.e. with a strong belief in the idea of ‘stocks for the long run’ and belief in things like the 4% rule. High stock allocations definitely work better on average, and they can be fine when coupled with flexibility to reduce spending (and avoid selling at losses) when the stock market is down. But for someone with a very long time horizon and with less flexibility for spending adjustments, this could be a recipe for problems.
The financial independence/early retirement movement does lean pretty heavily on the 4% rule. Do you agree that this is a safe withdrawal rate? Why or why not?
I am concerned about 4% being too high at the present for someone without much flexibility to reduce spending after market downturns. Reasons include: our extremely low interest rates plus high stock valuations have not been tested in the US historical data; the 4% rule has not worked internationally — only in the US and Canada but not in 18 other countries with data back to 1900; 30 years may not be long enough any more, especially for early retirees; it is hard for investors to earn the underlying index market returns net of fees.
[Reflecting further on the 4% rule’s historical context. For the 50 % stocks/50% bonds] portfolio, this was caused by the hypothetical retiree getting market returns from 1966-1995.
Over the whole historical period, the compounded real return on the 50/50 portfolio was 5.1% real
Over the 30 years, 1966-1995, the compounded real return on the 50/50 portfolio was 4.2% real (I call this market risk – it is less than the whole historical period).
But what was the return for the 1966 retiree? To know that, we have to calculate the internal rate of return from knowing that the 4% rule worked. That means $1 million supports 30 distributions of $40,000 plus inflation with nothing left after 30 years. That’s a compounded real return of 1.3%. That’s sequence risk! Somebody taking distributions from 1966 witnessed a much lower return than the overall market over those 30 years.
In light of that answer, and especially given the prolonged bull bond market, do you recommend a bond component in retirement portfolios? If so, what kinds of bonds, at what ages, in what percentages, etc.? And failing bonds, what’s your advice about annuities?
I think that bond funds are a very inefficient asset for meeting a retirement spending goal. As soon as someone wants to spend more for longer than the bond yield curve can support, this locks in failure for the plan. Alternatives include simple income annuities which are like bonds but with lifetime protection, or stocks. I think individual bonds are also valid since they eliminate interest rate risk when held to maturity to provide income.
The way I do simulations now accounts for lower interest rates in the short-term, and my estimate of the success rate for the 4% rule is about 70%, when keeping other assumptions about it like the 30 year retirement and historical equity premiums.
[As far as annuities pertain to FI/ER] I really haven’t dealt much with this for those in the 35-45 year old range. Conceptually, I think an income annuity could still be good at this point for someone not planning to work again, but I don’t know if there are practical problems with the pricing, etc., for someone that young.
If annuities are off the table, I think it makes sense to have a more conservative stock allocation around the retirement date with the idea to increase the stock allocation later. This can provide risk management for the sequence of returns risk in retirement.
Your advice and research tends to show more conservative results than that of your peers, such as Bengen and Kitces, and this has been the trend for years. While they are generally bullish on the 4% SWR, you’ve always advised more caution and lower WRs. Is this just a simple “different input = different output” issue? Or do you purposefully take a more conservative approach?
This is a quote I use from Bill Bengen to explain why he is comfortable with the historical numbers:
“This is hardly an exhaustive sample, and the case can clearly be made that no sample, no matter how large, would provide conclusive results since markets change over time. However, I maintain that the 87 years beginning with 1926 saw a wide range of market and economic conditions, including wars, depressions, oil shocks, “great moderations,” etc. Surely they contain something of value as precedent.”
I don’t know why I view this differently. It may just have been my starting place in looking at this area. I didn’t know a lot about the financial planning world until I wrote the article about how the 4% rule did not work nearly as well in other countries. I think we have to look beyond the US. And 4% is really vulnerable to small downward adjustments to return assumptions
In that regard, it may be partly about different inputs. I agree with their calculations about US historical data. But I just wish to look at a broader range of inputs than just US historical data.
If we’re decades out from retiring, how should we account for the increasing cost of healthcare?
I think this is a big problem that early retirees must address carefully. Medicare doesn’t kick in until 65, and right now there is so much uncertainty about what the costs of health insurance in the future will be. I don’t think there is any easy way to address this other than to remain flexible and to save a lot. There are calculators that can give you an idea about how much health costs will be over your lifetime, but they require a lot of assumptions that may or may not be right about future care costs.
I see that you’re very pro paying down mortgage ASAP. This is unusual to see, since a very common advice is to pay it off slowly and keep the cash in the stock market to come out with more money at the end of a 30-year period.
I’m not very pro paying down the mortgage. That’s just something I’ve started doing as stock market valuations get higher and higher. I’m still adding to my stock funds but just not as much as I otherwise might, because some is also being diverted to prepay the mortgage.
I’ve gone back and forth on my thinking about this. Mortgages are a way to leverage the investment portfolio and you ‘win’ if the market return is higher than the mortgage interest rate. You lose if market returns are less than the mortgage rate. I also psychologically don’t like having debt.
If we do see a big stock market correction, then I’d likely go back to reducing my mortgage payments so that I can go back to funneling more into the stock market.
Can you talk about why you use Monte Carlo simulation instead of historical? I find MC methods to be very useful for simulations in fields outside of market performance, but it seems like they miss a lot of critical trends like mean reversion. Are there any good ways to include those trends? To limit MC simulation to a more feasible space?
I think Monte Carlo and historical simulations can both be useful tools. But I like Monte Carlo because it makes it easier to incorporate things like lower interest rates in the short term, etc. The problem with historical simulations are that you are stuck with only the historical data and the data in the middle part of the historical period gets overweighted by showing up in more retirements than data at either end of the period. My preferred way to do Monte Carlo simulations now is to simulate bond yields and equity premiums and to construct bond returns and stock returns from these components. I simulate bond yields os that start of at today’s levels, are related over time, and mean revert (on average) to their historical numbers.
I’m curious as to your thoughts about how holding a global portfolio affects withdrawal rate calculations.
I’m a big fan of global diversification. Even if it doesn’t increase the returns, if it helps to reduce volatility then this is a way to increase the safe withdrawal rate (SWR). It’s hard to say the specific improvement because it requires assumptions about returns, volatilities, and correlations between all the asset classes. But as an example, for a case I looked at in my book, broader diversification increased the SWR estimate from 3.34% to 3.61%. WW1 & WW2 created some really bad SAFEMAX outcomes around the world, but even without that there are plenty of cases of withdrawal rates internationally falling well below 4%.
When equity valuations are as high as they are now, what type of investment/s would you recommend?
Have you won the game? i.e. have you saved enough that you are on a trajectory to meet your goals without additional market risk? If so, then start taking some of that risk off the table. If you are still far from retirement, maybe you can continue to withstand 100% stocks. Who knows if and when the market drop could happen.