In response to my post explaining our simple investment strategy of 90% stocks, 10% bonds, several readers inquired why we were in bonds at all. It’s a great question and one that I’ve been researching a bit over the past couple weeks.
As with most of our financial decisions, I try not to reinvent the wheel and would rather copy what smarter minds have come up with. The smarter mind I chose in this case was Vanguard, specifically the high-level allocation within their target date funds. While I chose the 2050 fund as an example in my other post, even their furthest out fund (2065 being the newest) with the most aggressive asset allocation also has the 10% bond allocation.
It turns out Vanguard’s target date funds have a glide path that maintains 10% bonds up until age 40 (with age 65 being the year in the fund’s name).
So, first I should try to figure out why Vanguard thinks 10% in bonds is a good idea and then try to see if that reasoning also applies to our own situation. Given our plan of retiring before 40 is a pretty big difference in goals from the average investor, maybe getting rid of bonds completely does makes sense. Let’s see if we can prove it one way or the other.
Why Does Vanguard have 10% bonds in their most aggressive target date fund?
First, let’s go to the Vanguard whitepaper discussing their approach to target date funds which can be found HERE. It’s a decently long read (packed with valuable information!), but I think I can boil it down into a few key points.
- Stocks have had higher average returns than bonds over all time horizons, but there is a higher chance of stocks significantly under-performing bonds in short time frames.
- Younger investors are able to withstand more risk because they have a large amount of “future earnings” in years ahead that is based on their ability to earn an income (“human capital”).
- Therefore, younger investors are better off investing the bulk of their capital in stocks, despite the potential short term risk, because of a long time horizon and stocks having higher expected returns.
- As investors get older, they should focus a higher percentage of their investments on bonds to protect themselves against the short term volatility, so the target date fund glides them into a more conservative allocation over time (see glide path chart above).
This is all pretty straightforward and sound reasoning, but doesn’t answer our underlying question of why 10% bonds for younger investors. For that we can dig a little deeper into the next section that compares Vanguard’s selected asset allocation to both a more aggressive (100% stocks to start) and more conservative (80% stocks to start) allocation.
The primary conclusion of this study was that the more aggressive allocation yielded higher average returns (as expected), but in the worst 5% of the simulations, the more aggressive allocation ended with a lower portfolio value. This is summarized in the following chart:
What I’m not a big fan of is that the two allocations compared simply move the equity allocation up/down 10% for all ages. What I suspect might be better is starting out at 100% stocks, then gliding into the 90% stocks at age 40 and following the actual allocation from there, but that’s purely speculation.
The next question is: Why did they stick to their own glide path given the results of this study? This leads to the more important philosophy behind the goal of the target date funds, providing sufficient retirement income. This is where some of Vanguard’s assumptions start to drift away from our early retirement dream:
- Contributions as a percent of salary (aka: savings rate) start at 5% for age 25 and drift up to 10% by age 65
- Retirement income needed is based off of 78% of age 65 salary
- Vanguard expects 46 of the 78% to come from Social Security
- The other 32 of the 78% is expected to come from private sources (aka: investments in the target date fund)
One way to provide this income is to take the lump sum of your investments and buy an annuity to guarantee an annual amount, which they estimate to cost 5.5x of the age 65 salary. The other way is to systematically draw down the fund as needed which has the potential to increase spending potential, but also introduces risk in the form of reduced spending potential.
Overall, they go into more detail on the percentages of success with the different methods, but never really give a definite answer as to why they decided on the asset allocation and glide path they did. The conclusion I drew was that Vanguard’s allocation choice has the same probability of success in the draw-down method as the more aggressive allocation, but has a slightly higher chance (1%) of coming out ahead in the annuity example.
One line that is really important as we “glide” into the next section of out analysis is this one:
“Over a longer period—through age 95—the investor’s probability of success is slightly higher on the more aggressive glide paths. For this reason, someone who decides to draw down his or her portfolio more heavily would benefit from such a path. For example, an investor who needs to replace 50% (instead of 32%) of his or her ending salary from private sources would have a 74% probability of sufficient wealth at age 95 on the more conservative path and a 78% and 81% probability on the Vanguard and more aggressive paths, respectively (not detailed in Figure 5).” – Vanguard’s approach to target-date funds
Vanguard’s Assumptions Versus Our Own
As mentioned above, Vanguard made several assumptions about the behavior of people investing in their target date funds that led to their selection of the asset allocation and glide path. Our own behavior is pretty far off from this:
- Vanguard assumes a modest 5-10% savings rate over 40 years.
- We are attempting to maintain a 50%+ savings rate over 10-15 years
- Vanguard assumes retirement spending is equal to 78% of income at retirement
- We don’t base our spending off of income (and you shouldn’t either!)
- Vanguard assumes ~60% of retirement income will come from Social Security
- We build our assumptions off of $0 from Social Security, so any extra will be a nice bonus in our 60’s
- Vanguard assumes ~40% of retirement income will come from personal investments
- We plan to fund our entire retirement (100%) from our investments
- Vanguard only projects retirement spending over 20-30 years after retirement
- If all goes according to plan, we’ll need to fund 50+ years of retirement if we exit before 40
Given how different our own situation is from the generic investor profile that Vanguard used to make their decision on asset allocation, it wouldn’t be surprising if our optimal asset allocation didn’t match theirs.
Vanguard’s Case for 100% Stocks
Vanguard recognizes that 100% stocks has the highest expected return when considering all combinations of stocks, bonds, and cash in a portfolio, and I even found a couple quotes to back up cases where Vanguard thinks 100% stocks is ideal for someone in our situation.
The first one I already mentioned above and comes from their Target Date whitepaper:
“Over a longer period—through age 95—the investor’s probability of success is slightly higher on the more aggressive glide paths. For this reason, someone who decides to draw down his or her portfolio more heavily would benefit from such a path. For example, an investor who needs to replace 50% (instead of 32%) of his or her ending salary from private sources would have a 74% probability of sufficient wealth at age 95 on the more conservative path and a 78% and 81% probability on the Vanguard and more aggressive paths, respectively (not detailed in Figure 5).” – Vanguard’s approach to target-date funds
The “longer period” they are referring to starts at age 65 and goes through 95, or 30 years. Since our goal is 50+ years, it appears having the higher portion of stocks leads to a higher chance of success.
The second one comes from another Vanguard paper on portfolio rebalancing:
“It is important to recognize that the goal of portfolio rebalancing is to minimize risk (tracking error) relative to a target asset allocation, rather than to maximize returns. If an investor’s portfolio can potentially hold either stocks or bonds, and the sole objective is to maximize return regardless of risk, then the investor should select a 100% equity portfolio.” – Best practices for portfolio rebalancing
I’ve heard an argument for holding bonds before that suggests having “dry powder” to invest in a potential market crash is worth the drag on returns. As Vanguard determined above, rebalancing on any regular schedule doesn’t have the ability to increase returns in a consistent fashion.
If you are holding bonds with the intention of converting some or all of them into stocks in a market downturn, then you are simply trying to time the market. Best of luck making the move at the perfect time to make up for the diminished returns from now until that point! If you’re really that good at timing the market (you’re not), then why not hold all bonds from now until the crash and convert them to stocks then? (This is a bad idea, please don’t do this!)
Time to Move Ourselves to 100% Stocks
Our investment strategy started out as copying Vanguard’s target date funds, but it turns out many of the assumptions that went into that asset allocation don’t apply to us. Specifically our larger than average savings rate that will allow us to retire early and live off of investments for over 50 years goes pretty far against the grain of the average investor the target date fund was set up for.
Migrating from 90% stocks and 10% bonds to 100% stocks isn’t a drastic shift, but it does simplify our holdings a bit. Rebalancing between US and international stocks will still be necessary, but based on the information in the Vanguard study on rebalancing portfolios linked above, this should probably be done yearly at most and only if there’s a shift our contributions can’t balance out.
This move will make our portfolio more aggressive, but in most cases should speed up our accumulation phase on the path to FI. In the low percentage of cases where it doesn’t thanks to a poorly timed market drop, we’ll simply have a work a little bit longer to make up the difference. A high chance of reducing our time working full time against a small chance of having to work a little longer seems like a good trade-off to make at this point in our lives. We’ll have to re-evaluate our asset allocation when we get ready to live off the investments and lock in our draw-down strategy, but for now 100% stocks seems to be the logical choice.
The most important consideration with an aggressive portfolio like this is being able to stick to the plan. If a 50% drop in the market would cause us to sell stocks for any reason, then all of the analysis above was pretty much useless. Now, the 90/10 split from before pretty much had the same problem. If we wouldn’t be able to handle a 50% drop in a 100% stock portfolio, then would we really feel that much differently in a 45% drop with the 90% stock portfolio? If you don’t trust yourself to stay the course when the market gets rough, then that’s possibly a more important consideration than sticking directly to the math laid out above.
Regardless of the market movement, 100% stocks is our new plan. Do you think we’re crazy or does our reasoning make sense? If you’re still holding bonds and are in a situation similar to ours, I’d love to hear why in the comments below! There’s always more to learn and I don’t pretend to have all the answers, maybe there’s something I missed.
Brilliant post. Here’s a “rule” that people have been accepting for decades and you question it by digging directly into the primary source – this is exactly the kind of critical thinking we need these days!
I fully agree with your analysis. I had come to the same conclusion after reading GoCurry’s 100% equity article but this is a fresh analysis that I really enjoyed. At the end of the day, I’d rather take the slightly higher risk of 100% stocks. I was holding ~10% too but it kind of felt like I was just appeasing all of the advice that says you need to have some bonds, i.e. “it feels like a good thing to do”. But as we know, we should never make financial decisions based on general feelings. Plus I know I can always keep working if I have to (and I have the extra comfort of decent family inheritance).
Thanks!
It’s always fun to challenge the “standard” way of thinking and dive into the actual data behind it. Definitely feels natural to have some bonds for general diversification, but 100% stocks seems to be the better long term choice. I agree that feelings have no place in optimal investing, but unfortunately they do cause people to lose money with poor market timing. It’s important to know yourself and invest accordingly.
The analysis is commendable. Not as much related to this post, but to the general reliance on cfiresim type tools: I believe the success rates are inflated.
Here is what I mean:
cfire calculates your success rate as number of paths your money lasts divided by number of paths it does not. The underlying assumption is that you “take the plunge” in a random year. This assumption is simply false. As the market valuation increases, your likelyhood to take the plunge increases (since you are hitting “the number”). High market valuations are connected with lower future returns.
Stated differently: You fail to calculate conditional probability of success (conditional on your reteriment year), and only relying on the base success rate, that can be vastly different.
Now, 100% stocks as your asset base would only exacerbate this bias. My point here is: you need an alternative measure of the portfolio value when you compare the portfolio value to your FI number. Otherwise, the variance in market returns makes you think you have “hit your number” when in fact you just experiences a “lucky” sequence of returns, that will turn unlucky the longer the lucky one was.
Basically, regression to the mean.
I think I agree with your general conclusion. People are most likely to reach their FI number during a growing market, and historical examples of a growing market would probably have a lower withdrawal success rate than looking at all historical markets.
Currently, as we are seeking to grow our investments with new contributions and additional market gains, I’m not too worried about the success rate of leaving at our FI number. For starters, we haven’t really declared a specific FI number as our spending is still variable year over year.
Second, I will be doing a lot more research on withdrawal strategies and the success rate of different allocations during the withdrawal phase of early retirement and FI. There’s a lot of variables in that, so for now I’m trying to maintain focus on optimizing the accumulation phase and will adjust accordingly down the line.
Thanks for the insight
One more profound idea I learned recently (that you can feel free to blog about) is that portfolio allocation should be related to other life risks.
Whenever we talk about investment risk we examine it in isolation – this makes no sense!
What we need to focus on is “life risk” – as in a whole picture of risks you face in life (including investment, marriage, wealth, job, etc). Basically, the point is to evaluate your risks as a whole and be able to trade off risks in one area for the risks in another.
Example: tenured professors have almost ZERO job risk – so they can afford a ton of variance in investment returns (yay stocks!)
Example2: independent contractors have significant job risks – they CANNOT afford a ton of variance in investment returns (so bonds might be a good idea).
While I agree with the general idea of considering “life risk” when choosing investment risk, I think the examples you provided are more akin to deciding how big of an emergency fund to have. Even someone with flexible or risky income should be focusing on returns in the accumulation phase, but should also have a plan to support themselves in the event of something unexpected happening.
Bonds can go down just like stocks, it’s often just not as severe. For a proper safety net in the event of income loss, some kind of cash or guaranteed return buffer is probably preferred. This prevents you from having to withdraw investments when they are down.
On the other hand, you have to compare it to how much money you will potentially lose over time by not having that money in higher returning assets. Basically, you shouldn’t spend $1,000 to insure yourself against a potential $800 loss if that makes sense.
Everyone’s situation is different, so determining the likelihood of needing to withdraw money from investments (and how much) is important when determining asset allocation (including cash in a savings account).
Great post Noah! I have been learning about these same things myself in the last few months and you write things out so clearly- it helps me to feel solid in some of the same ideas I am inclined to myself but couldn’t spell out so clearly. My circumstances are much more complex, as I am balancing for myself, my 66 year old mother, and college planning for my 12 year old son, all as a low income single mother. But basically I am feeling like this- a good cash back bank account can produce better yields than bonds and seems a much better way to hedge the influxes of the market. As we will need continual withdrawals, but in lower amounts, if we have 3-7 years of needed funds in bank accounts then we can weather most stock market fluctuations- they all turn around within 7 years, if they don’t, we would have been screwed with bonds too and the whole nation is probably screwed with us. Since our interest rate would beat the bond rates, we’re not loosing anything. The rest can be considered long range monies and in things that earn best for long range ,with periodic withdrawals to maintain our 3-7 year cash holdings. Then separately is college $, which need to be invested in something that can be cashed out in 3 years and put in a 529 plan for Fafsa applications. While I really focus on growth and hate watching investments behave like snails, I must balance that I can’t risk these assets taking a plunge and then needing to cash them all out “low.” And then there are the further life circumstances that really monkey up the balancing- knowing that there is a heavy possibility that I may have to sell everything at anytime to settle a family real estate dispute changes all the balances- keeps me needing to stay very conservative for now and invest as I would normally later. Even though I hate those snaily bonds. I looked at 10 years of history on these investments and there are some clear winners and loosers, over any time range- and I’d sure love to ditch those loosers. Guess I might just replace at least some them with bank accounts…
Thanks again for clarifying this aged misnomer!!!
Hey Lela, thanks for reading!
The major concern you seem to have brought up is the performance of bonds versus a savings account. What interest rate are you getting on 100% guaranteed cash that seems to be beating the bonds? It’s easy to find accounts ~1%, but getting beyond that often takes a good amount of work, especially if you’re talking about 3+ years of expenses it might not even be possible to get above 1%.
In that case, bonds will almost certainly have a higher return. Are you investing in a broad diversified bond fund like VBTLX or attempting to pick individual bonds? I would highly recommend not picking individual bonds, but your comment about ditching losers has me concerned. The fund I linked has averaged over 2% for the last ~5 years and even higher than that looking at the last 10 years (although forecasts do have future yields a bit lower). A bond fund like this will almost certainly beat inflation whereas your savings account will not.
A 529 can hold investments and provide tax benefits on the gains, so if you know money will end up in one, it’s probably best to just put it in there now. Cashing out something later and then adding to a 529 will probably end up with higher taxes paid. I agree this allocation should be conservative given a 3-year time frame, but you should be looking at the time frame when the money will be used for college, not when you plan to put it into the 529.
Back to your plan of holding 3-7 years of expenses in cash, you’d have to run the numbers but this seems like a terrible plan in most situations. Protecting against what you describe as a worst-case 7 year market downturn is extremely expensive in lost returns compared to what you might pay by having to pull some funds out when the market is down. I agree your situation may call for a larger than average emergency fund, but anything beyond a year or so seems excessive. You’re putting a large amount of money at risk in regards to future inflation that wouldn’t be the case in stocks or bonds (in general). You’re so worried about losing money in investments that you are potentially setting yourself up for a much higher chance of overall investment failure.
If you want to see this for yourself, I recommend playing around with cFireSim or FIREcalc and running simulations with a decent % of your funds allocated to cash at any given time. I believe even the trinity study that looks at 30 year retirements covers the dangers of holding cash iirc.
http://www.cfiresim.com/
http://www.firecalc.com/
Another good resource is this Vanguard article on asset allocation, specifically the section titled: “Stocks are risky—and so is avoiding them”