Is tilting index funds ever a good idea?

[alex123711]

Couldn't really word the title right, but is tilting such as small cap, small cap value (SCV), mid cap or mid cap value (MCV) etc ever a good idea? SVC and MCV have outperformed previously, though who knows if this will continue, the reasons SCV outperformed seem to make sense to me.

[Crusader]

It makes sense. But the whole factor model seems to me like a bit of a data mining exercise, I am not fully onboard (which doesn't mean I won't be, one day). In the end I decided against tilting my portfolio because of my hesitation and for simplicity's sake. If I was to tilt it, I would use AVUV.

[Scott 2]

Have you read about factor based investing? This may be what you are thinking of:

https://www.amazon.com/Your-Complete-Gu ... 0692783652

[Western Red Cedar]

John Bogle on tilting:

"Impressed both by the long-term performance (and recent performance) of value stocks and small-cap stocks, some investors hold the all-market (or S&P 500) index fund as the core, and add a value index fund and a small-cap index fund as satellites. I'm skeptical that any kind of superior performance will endure forever. (Nothing does!) But if you disagree, it would not be unreasonable to hold, say, 85 percent in the core, another 10 percent in value, and another 5 percent in small cap. But doing so increases the risk that your return will fall short of the market's return, so don't push too far."

-- The Little Book Of Common Sense Investing, p.206.

[jacob]

Pro-tip: You can run any number of analyses on your data set (e.g. price time series ... and maybe something else). Factor analysis is the simplest kind. After that comes cointegration and principal component analysis. What really matters is whether you can attach whatever your analysis picks up to some narrative---a particular investor-type. For example, P/B and smallcap as factors make sense as the former is driven by the value investor demographic while the latter is the playground of non-institutional analysts (hedge funds and "small-time" 7-8 fig operators). If you can't you're just data dredging, as mentioned above.

If you had come up with this idea and published it in in 1992, you would have won the Nobel two decades later. See French/Fama.

It may be a good idea insofar you're one of the few who have identified this before the result became widely known. In order for this to work, other people would need to realize later that it really is a good idea. It would also be a good idea even after the fact or in spite of the non-realization by intentional investors if it's based on some permanent psychological narrative, that is, humans doing human w/o a deliberate strategy.

[Lemur]

I tilted my portfolio this year on large cap value (VTV) after noticing that the total stock market is down ~16% YTD but large cap value has held most of its value at only down 1%. So I wanted my equity mix to tilt towards value specifically because I believe I can still get growth but for less risk in the long-run. Its likely I'm already late to the party doing this. Big money flocks towards value when PEs and Tech start looking grossly overvalued or they're chasing more reliable equities when the market is in a downturn or any other number of negative macroeconomic situations.

Will this outperform the market in the long run? I do not know for sure but outperformance is not the goal - it is just one part of an allocation for a portfolio mix that I feel I can stick to and sleep well knowing there is some data supporting this. And there should be some positive performance here as long as the market has booms and busts, rises and falls, and fluctuating media headlines between whether the Oracle of Omaha is past his best days or was he ahead of the curve this whole time?

I suppose this argument could potentially extend to other [insert index fund here that tracks this] due to [outperformance during this time frame]. A portfolio mix is really just ones own personal preference of expected risk-rewards...like a dietary preference perhaps. So to answer the question of is it ever a good idea- I think the answer is it depends.

[alex123711]

I think tilting was probably the wrong word, I was more talking about an allocation towards say SCV and keeping that allocation as opposed to changing the allocation depending on other factors, although that is also an option.

[ertyu]

If you stick with an allocation without changing it in response to external economic circumstances it is best to choose large cap/total stock market. This is because of how the US retirement system currently works. As long as there are no changes to the retirement system, there will be somewhat predictable, consistent, and price-blind inflows into the total stock market index as people save for retirement. During recessions, these flows might be less, but they will still arrive on a regular schedule as people dollar cost average their savings into "the stock market." There wont' necessarily be such consistent, price-blind inflows when it comes to small caps/value.

[Crusader]

What is the source of this line of thinking?

[ertyu]

What is the source of this line of thinking?

mike green, simplify asset mgt. does podcasts, used to manage peter thiel's money. can't link to a precise podcast but he's the guy that alerted me to the importance of retirement flows

[jacob]

If you stick with an allocation without changing it in response to external economic circumstances it is best to choose large cap/total stock market. This is because of how the US retirement system currently works. As long as there are no changes to the retirement system, there will be somewhat predictable, consistent, and price-blind inflows into the total stock market index as people save for retirement. During recessions, these flows might be less, but they will still arrive on a regular schedule as people dollar cost average their savings into "the stock market." There wont' necessarily be such consistent, price-blind inflows when it comes to small caps/value.

Also see https://earlyretirementextreme.com/the- ... sting.html

[Crusader]

Well, shit. I am now not as sure of my portfolio (I am an index investor). Jacob, I was aware of the problem of how people calculate market capitalization, and it always bothered me. It's a dynamic/instantaneous price, you can't just take it and multiply by the number of shares. I'll have to think about this some more.

However, being in it for the long term does not make investing in stock risk free. If this was the case, institutional investors would run arbitrage and bring returns down.

I don't understand what you mean/how they would run arbitrage. Would you mind elaborating?

[jacob]

I don't understand what you mean/how they would run arbitrage. Would you mind elaborating?

Insofar the "risk of equities decrease in the long run", arbitragers would sell risk-free long term bonds and buy equities for holding long term(which should have more reward, because more "risk"). This will bring down the price of bonds (yields go up) while bringing the price of equities up (ROI down) until the yields are the same.

This is not happening.

The reason this trips people up and leads to long and boring conversations is that "risk" is ill-defined. People tend to speak of it as a single number (and associate it with a reward, as least within MPT, but it really is a range of numbers. E.g. equity has a higher average return than bonds but also a wider range of potential returns some of which are lower than the bond average (different range). This is why the arbitrage is not happening.

[Crusader]

Ok, I see what you mean, but I am not sure I agree. For all intents and purposes, I think of "risk" as the variability of returns over a fixed time period. If you are the type of person that can tolerate a range of returns (or range of time periods, over which to retire, for example), you might want to buy that "riskier" stock. You are getting a premium on the price for this variability, though. In other words, I think the riskier and less-risky assets are priced properly. On the other hand, people who are very sensitive to left-tail-risk might want to buy more bonds.

I don't know how institutions invest and what their objectives or risk tolerances are, so I don't know why they don't buy more stocks and less bonds. I imagine an institution has to balance that inflow and outflow of cash that you talked about with the FIFO stack, which makes them less risk-averse.

[white belt]

Ok, I see what you mean, but I am not sure I agree. For all intents and purposes, I think of "risk" as the variability of returns over a fixed time period. If you are the type of person that can tolerate a range of returns (or range of time periods, over which to retire, for example), you might want to buy that "riskier" stock. You are getting a premium on the price for this variability, though. In other words, I think the riskier and less-risky assets are priced properly. On the other hand, people who are very sensitive to left-tail-risk might want to buy more bonds.

On the surface that makes sense, however there are a lot of underlying assumptions built in to make that true (Jacob points out the demographic assumptions in his article):

1. The market is properly pricing in all risk (Efficient Market Hypothesis)
2. Variability in returns has the same impact on all market participants. If I'm a long-only retail investor, why do I consider upside risk a risk at all?
3. Bonds are less risky than stocks. There are a number of macroeconomic factors that can render this untrue to include inflation, government debt, etc. Look at past year to see how bonds and stocks are correlated in certain macro environments.

Jacob is pushing back (and has been for 15+ years) on the widely held belief in certain circles that in the long term, stocks can only go up and/or are less risky than bonds. It's worth understanding all the unspoken assumptions that have to hold true to keep that belief afloat. This will allow you to determine if/when those assumptions no longer hold true and a change in strategy is warranted.

[white belt]

I think tilting was probably the wrong word, I was more talking about an allocation towards say SCV and keeping that allocation as opposed to changing the allocation depending on other factors, although that is also an option.

So if I'm understanding correctly, you are essentially asking if active investing makes sense? Of course the tongue in cheek answer is "yes, but only if you can generate Alpha."

You can make allocation decisions based on all sorts of models, but I would first clearly define what your strategy is. Certain sectors will outperform at certain times, but to make money on such knowledge requires the following:

1. You can identify the underlying conditions that cause certain sectors to outperform
2. You can predict when those underlying conditions will occur
3. You can position ahead of other market participants who have the same idea (maybe because they are using the same models/datasets you are, but more likely they have even better models/datasets)
4. You can exit the position at a profit prior to when the rest of market participants attempt to exit their positions
5. You have the emotional/psychological fortitude to execute steps 3 and 4

The above is true regardless of if your time horizon is measured in hours or years. Keeping an allocation is making an active trading decision just as much as changing an allocation is.

[WFJ]

Simple answer, Ex-post = Yes, Ex-Ante = No.

Minimize fees, minimize taxes, understand portfolio risks and cash flows are much more worthwhile activities to pursue in order to maximize long term market returns. The strategy to buy VOO and spend your time doing something will always dominate any market tilting strategy over the long run.

[Crusader]

On the surface that makes sense, however there are a lot of underlying assumptions built in to make that true

When you reference "that", I am not really sure what you mean by "that". Jacob essentially said that if the usual theory is true, there would be no reason for institutional investors not to just buy more stocks and forget about bonds, because why wouldn't they if on average stocks always yield more returns? Since that's happening, something doesn't quite add up. And I am saying that there still is a reason for institutional investors to buy bonds, precisely because they might need to reduce variability (for whatever reason... maybe they need to prepare for a bunch of withdrawals).

I don't think that anyone is claiming that bonds are not "risky", they are just less likely to be risky compared to stocks. I mean of course the interest rate, for example, will directly affect the price of bonds.

[white belt]

When you reference "that", I am not really sure what you mean by "that".

"That" is your entire definition of risk, presumption that risky assets are priced appropriately, and presumption that there is a linear relationship between risk/returns. You are referencing the CAPM model, which comes with many flawed assumptions that have already been pointed out by Fama-French and others 30+ years ago.

[alex123711]

If you stick with an allocation without changing it in response to external economic circumstances it is best to choose large cap/total stock market. This is because of how the US retirement system currently works. As long as there are no changes to the retirement system, there will be somewhat predictable, consistent, and price-blind inflows into the total stock market index as people save for retirement. During recessions, these flows might be less, but they will still arrive on a regular schedule as people dollar cost average their savings into "the stock market." There wont' necessarily be such consistent, price-blind inflows when it comes to small caps/value.

I am not in the U.S