Hi everybody! I’m back from a two-week blogging hiatus! Things got busy at the office right before I left and we also had to prepare for our road trip and ERN Family World Tour, currently in beautiful New Mexico and moving on to Texas soon! I was amazed at how little work I got done while traveling! Early retirement is a lot more work than I thought!
In any case, today’s topic has been on my mind for a while: What would be reasons to hold individual stocks? Not all but the majority of folks in the FIRE community apparently favor just plain passive index investing and I have been an index investor myself for the longest time. But occasionally we should definitely question our assumptions. Especially those that sound like the good old “We’ve always done it this way!” And one “excuse” to look into this topic is the ChooseFI podcast featuring Brian Feroldi a few weeks ago. Brian talked about his adventures as a stock picker! I thought it was a great episode, though, of course, I didn’t agree with everything. But it got me thinking about what would be good reasons and what would be not so good reasons for me to abandon my index-only approach. Let’s look at my favorite eight…
Bad reason #1: I’ll beat the market
As many smart people before me have pointed out, all active stock pickers in aggregate (!) can only at most generate the market return. That’s because the market overall has exactly the cap weights (by definition) and when you take out all the passive indexers (who also hold exactly the index weights, by definition) then what’s left for all the non-passive investors is also distributed exactly according to the index weights. Take out the cost (both time and monetary) for researching stocks and the stock pickers will likely lag behind. Again, at least in aggregate!
But even among the stock pickers that – against the odds – manage to beat the market, the outperformance may not be due to skill at all. Here would be three reasons why someone could have outperformed the market with no skill involved whatsoever:
- Luck: Ten monkeys throwing darts at a board with stock tickers will, on average, match the index. Around five of them will beat the market and the remaining monkeys will underperform. Statistical tools are necessary to sort out whether someone was skilled or just plain lucky.
- Sector and/or style bias: Lots of investors beat the market all the time. I’m sure that everybody who was heavily invested in IT stocks would have outperformed the market over the last few years. So, imagine the IT sector outperformed the market by 5% and a stock picker who’s heavily invested IT outperforms the market by only 4%. Would I call that person good stock picker? Probably not! You could have just bought a sector ETF and done better! Similar issues arise when investors take other significant departures from the market weights, e.g., value/growth tilt, small-cap bias, etc.
- Beta bias: Everyone can beat the market over the long-haul. Just use leverage! One can use leverage explicitly, which is quite easy and inexpensive if implemented with futures (see this old vintage post of mine on the Synthetic Roth IRA). Or by overweighting “high-beta” stocks, i.e., stocks that are more responsive to market moves than the average. So, if someone has a portfolio beta of 1.4 (40% more market risk than the broad index) but less than 1.4-times the market return it wouldn’t be skill at all despite outperforming the index! As we like to say in finance: outperformance due to beta ain’t alpha!
So, if people tell me that they beat the market I would normally offer to run a statistical analysis to quantify these three potential sources of outperformance. Without getting too much into the finance weeds, it’s called a factor model and it’s a standard tool in finance to (among other things) sort out exactly this question: determine whether there is any “alpha” (=outperformance) left after accounting for beta and commonly used styles (often value and small-cap, but the academic literature has come up with many more factors). And if there’s any outperformance, how statistically significant is it? So far, no one has taken me up on that offer. Maybe Brian wants to volunteer?! Send me your monthly %-returns and I’ll start cranking! 🙂
Bad reason #2: Lower fees
True, you will save the expense ratio of the ETF or mutual fund. But how much money are we talking about here? A $1,000,000 portfolio invested in index ETFs/mutual funds at Schwab (SCHB/SCHX), Fidelity (FSTVX/FUSVX), or Vanguard (VTI/VOO) will cost you $300, $350, and $400 a year in expenses for the standard broad U.S. equity index funds (both Total U.S. Stock Market and U.S. Large Cap).
However, once you factor in the time you spend on researching stocks (opportunity cost), the cost of newsletter subscriptions, trade commissions, etc., I doubt that you’ll come out ahead of the index investor (but note the caveat in “Good Reason #3” below!).
Bad Reason #3: Warren Buffett did it!
Warren Buffett seems to be the only stock picker not only spared the ridicule of the FI community but he’s even revered by us. And for a good reason because Berkshire Hathaway has definitely had a great run of outperforming the overall market. Then why do I call this Bad Reason #3 and not Good Reason #1? Well, here’s what researchers at a pretty well-known asset manager did: they ran a factor model similar to the one I mentioned above and found that a sizable portion of the Berkshire-Hathaway outperformance is due to leverage:
“One of the ways that Berkshire Hathaway was able to add so much return above that of the market is Berkshire’s access to cheap leverage via its insurance business, allowing it to harvest greater amounts of these style exposures than most traditional investors could.”
Out of the over 10 percentage point of excess return over the market portfolio, only 3.6% are due to “alpha” (skill) and that regression coefficient is also not even statistically significant. For us retail investors it means that without the benefit of owning an insurance company on the side we likely wouldn’t be able to get even close to replicating what Warran Buffett did. Even if we had done all the same work and brilliant analysis as Buffett and were able to hold those same stocks (which is also impossible because of some of the private, non-listed holdings)! The best way to replicate what Buffett is doing is to simply hold BRK-B in our portfolio.
Bad reason #4: I will invest only in stocks I know
I think that’s a horrible reason. Most folks following that strategy will inadvertently introduce a significant bias into their portfolio. You may know certain stocks very well because of the industry you work in. But I would have never invested in financial equities beyond their index weight because they are highly correlated with my wage income. Others hold companies that they like to frequent as consumers. You’ll likely introduce a significant bias to consumer and IT stocks and are underweight Industrials and Materials. Even if you manage to outperform the index you will probably do so at the cost of much higher volatility! In any case, I looked through the list of S&P500 stocks (there are actually 505 of them, go figure!) and there were tons of stocks that I never heard of before. And that’s for the 500 largest stocks. If we go to mid-cap or even small-caps I would probably know hardly any of them! And I love it: I hold stocks I didn’t even know existed! You can’t get much more diversified than that!
So much for the bad news. Let’s move on to four good reasons!
Good reason #1: Tax loss harvesting
I like Tax Loss Harvesting. It’s a neat little the tool that may add a few extra basis points to the after-tax return. And it’s more effective if one can do that on the individual stock level, not just on the stock index level. Look at the following simple example to illustrate this point. Imagine, for simplicity, an index has 5 stocks and they are equally weighted. Imagine I have two options: invest $1,000 in each of the individual stocks or $5,000 in the index. Imagine also that there is some dispersion across the five stocks. Specifically, the individual returns are -4%, -2%, +/-0%, +2% and +4% in addition to the median stock (=index) and of course no one knows in advance which one of the five stocks will under or outperform the index. So, if we calculate how much in tax losses we can harvest as a function of the index return between -5% and +5% we notice that holding the individual securities generates more in tax losses. All the while the pre-tax returns are the same.
The biggest difference occurs when the index is flat. In that case, there is clearly no tax loss for the index investor but the stock investor can still harvest the losses of the two “underwater” stocks! Holding individual stocks will generate more tax losses unless we find ourselves in one of the two extreme events where either all stocks underwater or all stocks gained!
And for full disclosure, this is not an endorsement for any of the (for-fee) Robo-advisers. Quite the opposite, the few basis points in extra return don’t justify the 25 basis point (0.25%) annual fees that Wealthfront and Betterment charge. You will never find an affiliate link for those guys on my blog! Instead, just do the tax loss harvesting yourself with only a minimal amount of work, check my blog post from a long time ago on how to be your own Robo-adviser!
Good Reason #2: More tax-hacking
In my last post from two weeks ago, I pointed out one little “flaw” in the index-only investing strategy: investors in a high enough tax bracket will be hit with taxes on dividends in the accumulation phase: 15-20% marginal on the federal level plus 3.8% Obamacare tax plus state tax (if applicable). Those taxes are assessed every year and compound over the years! So, even if you don’t like to deviate from the index weights you could benefit from constructing your own U.S. large-cap index portfolio from scratch and then putting the lowest-yielding stocks in the taxable account and the highest yielding stocks in the tax-deferred accounts. One would then avoid compounding the dividend tax bills in taxable accounts. I proposed that this could be easily implemented with an ETF but until Vanguard, Fidelity, Schwab, iShares, and State Street step up to the plate one would have to do this manually with individual stocks!
Good reason #3: It’s fun and it will keep me engaged and motivated
Investing is for the long-run. Even early retirement will take 10-12, maybe 15 years to achieve for most of us. How do you get excited enough to start this journey and how do you stay focused and motivated along the way? I have the impression that for a lot of FI bloggers, investing is mostly a means to an end. Investing is boring! There’s even a GoCurryCracker blog post with that title. I have no problem if people want to make investing boring. But I think that there are many among us that are truly excited about investing. I would count myself in that group, though I’m more excited about investing from a macro perspective, less about stock picking. But if Brian wants to subscribe to a stock newsletter, more power to him! He seems to really enjoy reading updates about companies. It’s like Brian is reading People Magazine, but with stocks! Can’t we all agree that this a more productive use of his time than reading the actual People magazine?
And here’s another example: how will I get my daughter excited about investing? She’s only four right now so this discussion is still a few years in the future. If the time comes in a few years then, sure, I could tell her “Hey, kiddo, I bought you one shiny share of VTSMX and sorry, VTSAX is only available for accounts $10,000 and up!” Maybe she’ll get really excited about that. But I suspect a Total U.S. Stock Market Fund is a bit too abstract for a young kid. Compare that with giving her a share of Disney or McDonald’s, two companies she really, really likes even now at only four years old (God bless America!). Maybe even give her the actual stock certificates, framed to hang in her room (even though I heard that getting the physical certificates is not easy anymore). That might be a lot more impactful for a kid!
Good Reason #4: Some prefer a style that requires weights different from the index cap weights
OK, so now we are wading into territory that might spark some controversy. But here would be a few reasons to deviate from the market cap weights and the implementation may (or may not) involve holding individual stocks:
- Dividend-tilt: Overweight stocks with a higher dividend yield, for example, to generate income in (early) retirement. Some bloggers swear by it (Ten Factorial Rocks, Tawcan, etc.) and I must say I’m also warming up to the idea. Some of the high-dividend-yield funds get close to or even match the 3.25-3.50% withdrawal rate I’m targeting. The thought of not having to dig into principal during early retirement is certainly appealing!
- Sector tilt: Overweight and underweight certain sectors or even certain individual stocks. For example, if you work in a specific sector or for a specific corporation, you under-weight that specific stock/sector so as to not double up on the sector risk or idiosyncratic company risk.
- Beta tilt: As a risk-control, one could overweight stocks that have “low beta,” that is, lower exposure to market risk. There’s even a very well-known research paper, aptly titled “Betting Against Beta” that showed that low-beta stocks tend to have better risk-adjusted returns. And the same phenomenon also holds in other asset classes!
- Equal-weighting (as opposed to market cap weighting): Past returns show that equal-weighting would have outperformed the cap-weighted index. See, of all places(!!!), a link at Bogelheads, with returns going back to 1971. What I like about equal-weighting is not even so much the past outperformance because there’s no guarantee that this will carry over into the future. But the risk, specifically the idiosyncratic equity risk, is spread out more equally than with market cap weights, where the top 50 stocks in the S&P500 account for only 10% of the names but 50% of the market capitalization!
Admittedly, some of these can already be harvested through ETFs. For example, there are a number of High-Dividend ETFs or sector ETFs. Equal-weight ETFs have been around for a while as well. But the more exotic the ETF the higher the fees, for example, 0.35% for the “Dividend Aristocrats” (NOBL) and 0.30%+ for most equal-weighted ETFs. That could already entice some of us to simply replicate the whole thing with a number of individual stocks. Also, to my knowledge, ETFs haven’t picked up the exact “betting against beta” theme yet, though some ETFs exist in a related space: Vanguard, Fidelity and iShares all have “‘low-volatility” ETFs. While vol and beta are related there’s a subtle difference between the two, of course!
I’ve been an index investor my entire life. But I will explore investing in individual stocks going forward. Not necessarily because I have any illusion that I’ll be the next Warren Buffett but mostly because I like to shift some of our equity portfolio into the low-beta, high dividend and equal-weighted direction and do some additional tax loss harvesting. To keep transactions costs as low as possible I will likely use M1 Finance (affiliate link), a new online brokerage that offers commission-free trades and trading fractional shares of equities and ETFs! Stay tuned for updates on how this goes!