How to 'Juice' an index?

[ILikeDividends]

I have a feeling this forum is heavily biased toward indexing, so hopefully this hypothetical scenario won't be too heretical.

Suppose you have $x to put to work. Say you put 80% of x to work into an equivalent of SPY (in Schwab's case, that would be SCHX, market cap weighted, low expenses, no transaction fees).

You take the remaining 20% of x and spread it across the top 10 holdings in SCHX.

Off the top, that 20% would incur some transaction fees that you wouldn't otherwise incur by putting 100% in the index.

In your opinion, what would be the likelihood that you could marginally outperform the index, overall, even after accounting for the additional transaction fees in the individual issues?

For extra credit, should that 20% be allocated across those 10 issues by market cap (like the index) or equal weight?

[ketchup]

Surely simple large cap index funds exist that you could just buy on top of a whole market index if you want to emphasize that in your portfolio?

[ILikeDividends]

Surely simple large cap index funds exist that you could just buy on top of a whole market index if you want to emphasize that in your portfolio?

Excellent point.  XLG has the 50 largest S&P 500 issues.  A little more broadly based than I was thinking about, though.  On the plus side, it's also transaction free on Schwab.  But a $4 spread between ask and bid could mitigate the lack of a transaction fee. :(

[Radagast]

So far as I know the top ten stocks are historically more likely to be the under-performers going forward. You'd be better off selecting the middle 10, last 10, or a random 10. In fact randomly selecting 100, holding forever, and eschewing the index fund is supposed to be a pretty good choice. It will have lower costs within a few years, and probably better performance.

[Indexer]

I would actually go in the opposite direction. Instead of over weighting the top companies maybe you should seek further diversification with small caps.

SPY is 500 companies. VTSAX is around 3500 companies, and the top 500 are the same as SPY. You are getting more diversification at a lower cost. In addition, over the long term small caps tend to outperform large caps. The ETF version is VTI.

Side note: SPY is cheap, but for what it is it is incredibly expensive. 0.09% for SPY. VOO is the same thing and costs 0.04%. ;-)

[Rubic]

Agree with the other posters.  If I wanted to skew outside the normal S&P 500,
I would probably allocate toward an equal-weight ETF such as the Guggenheim RSP.

[sokoloff]

Surely simple large cap index funds exist that you could just buy on top of a whole market index if you want to emphasize that in your portfolio?

Excellent point.  XLG has the 50 largest S&P 500 issues.  A little more broadly based than I was thinking about, though.  On the plus side, it's also transaction free on Schwab.  But a $4 spread between ask and bid could mitigate the lack of a transaction fee. :(

You might have been looking at an after-hours spread. Looking now, with the market open, I see a $0.16/share spread (a little under 9 basis points).

[ChpBstrd]

over the long term small caps tend to outperform large caps. The ETF version is VTI.

To see this point, pull up a long-term chart of VB (small caps) versus VTI (total market). But also note how VB is more volatile. If a bear market develops, a switch to small-caps might be regrettable.

The best way I know to beat the market is to sell a put option ITM and get assigned. Then you have the option premium received plus market performance. For extra credit, invest the premium received.

E.g. At this moment, $259 December 15 SPY put options can be sold for $4.11 per share (or $411 per contract). If you get assigned, you've essentially bought SPY for (259 - 4.11=) $254.89 at a moment when the market price was $257.60, a full 1% discount. Hold those shares for a year and you will have outperformed the person who just bought the index by 1%, minus transaction costs/taxes, regardless of if the market rises or falls.

Note that with this strategy there is a risk the market surges higher in the next few weeks, and your put expires out-of-the money, leaving you with $411 in your pocket, but no shares. That's the risk you're getting paid 1% in 35 days for, and in that event you would play again until assigned, earning ~1% per month until eventually getting assigned on a dip.

Side note: SPY is cheap, but for what it is it is incredibly expensive. 0.09% for SPY. VOO is the same thing and costs 0.04%. ;-)

The difference may be that for whatever reason, VOO has a thin, relatively illiquid options market with $0.60 wide bid-ask spreads for next month's options (!) and few available strikes. SPY, in contrast, has dozens of strikes and spreads of 1-3 pennies per contract. The five one hundredths of a percent difference in fees would be quickly made up on the first trade I described above. For example, the same strategy played on VOO at the $240 strike (the only December ITM strike available) and selling at the midpoint of the bid-ask would yield only 0.3%, which is not worth the trouble. Note how the 0.7% net difference in put-purchase-strategy outcomes would cover 14 years of the difference in fees between SPY and VOO.

This has always struck me as a paradox. SPY is the bigger and more popular ETF despite higher fees. Because it is bigger and more popular, it has the better options market. Because it has the better options market, it is bigger and more popular.

Bottom line: if you're just clicking "buy", VOO is the better investment. If you purchase with puts as described above, definitely use SPY.

[Financial.Velociraptor]

If your index has options, you could sell covered calls for some extra income.  You potentially give up some upside but you lower your risk and keep the income win/lose/draw.

[Retire-Canada]

In your opinion, what would be the likelihood that you could marginally outperform the index, overall, even after accounting for the additional transaction fees in the individual issues?

0% likelihood.

[Systems101]

In your opinion, what would be the likelihood that you could marginally outperform the index, overall, even after accounting for the additional transaction fees in the individual issues?

0% likelihood.

+1

Go put the 20% in SCHA

[ChpBstrd]

In your opinion, what would be the likelihood that you could marginally outperform the index, overall, even after accounting for the additional transaction fees in the individual issues?

0% likelihood.

This is not the case if you pursue a put-assignment or covered-call strategy against a market-representative index. In those cases, you are trading cash in hand for some risk of underperforming the market. Thus, compared to buy-and-hold, you have taken on additional risk in exchange for additional return. There is some level of probability that the risk will not occur, and that you could outperform the buy-and-hold strategy. That level of probability is the answer to the OP's question. There's even software that can calculate the odds of outperforming the index (option expiring ITM) for you.

[Retire-Canada]

This is not the case if you pursue a put-assignment or covered-call strategy against a market-representative index.

And is that what the OP was suggesting doing in the post I quoted? No. Hence my reply.

[ChpBstrd]

This is not the case if you pursue a put-assignment or covered-call strategy against a market-representative index.

And is that what the OP was suggesting doing in the post I quoted? No. Hence my reply.

I agree I'm not talking about the OP's original proposal, which is a bad idea.

However, even with the OP's stock-picking / sector-picking idea, there is some chance of outperforming the market. For an extreme example to illustrate the point, consider if the OP asked the odds of beating the market playing casino roulette and betting on black in one roll with his entire portfolio. Odds of outperformance = 49% Risk = astronomical. Similarly, there's some low chance a stock/sector/theme picker will happen to select the next 10-bagger. So the odds aren't 0%.

There are lots of opportunities, like the roulette wheel, to get outsized returns, but these are rarely a good idea in risk-adjusted terms. I.e. a 49% chance at double-or-nothing is a gamble no mathematically-aware person should take. I think the OP's proposed strategy is only a marginal improvement because it has lower downsides than roulette!

[Retire-Canada]

I agree I'm not talking about the OP's original proposal, which is a bad idea.

The chances of the OP's original proposal generating an improved return after fees may not be precisely zero, but would round to zero.

[Proud Foot]

You take the remaining 20% of x and spread it across the top 10 holdings in SCHX.

Off the top, that 20% would incur some transaction fees that you wouldn't otherwise incur by putting 100% in the index.

In your opinion, what would be the likelihood that you could marginally outperform the index, overall, even after accounting for the additional transaction fees in the individual issues?

For extra credit, should that 20% be allocated across those 10 issues by market cap (like the index) or equal weight?

The biggest thing would be your transaction costs and whether you would adjust those 10 issues as the index changes. As far as your question you could outperform the market. Using the current top 10 (11 as I used both GOOG and GOOGL so it would be the top 10 companies) you would have outperformed if you invested in those in June 2013 at the FB IPO.

Using an equal weighting of 2% (1% each GOOG and GOOGL) gives you a CAGR of 17.49% vs 15.43% for SCHX. However a portfolio of only those individual issues and the same weighting returned a CAGR of 25.64%. All companies returned a CAGR higher than SCHX with the exception of Exxon Mobil (4.21%).

However that is based upon what is currently in the top 10 and there is no guarantee of matching those returns going forward. The Top 10 holdings in 2012 used in your strategy had a CAGR of 14.6% and the individual issues had a CAGR of 11.16%. Only 4 of the current top 10 were in the top 10 in 2012 (Apple, Microsoft, Johnson & Johnson, and Exxon Mobil)

[Scandium]

Why do people still buy the S&P? Schwab has SCHB for 1 bp more. Thousands of stocks vs 500 large cap only. Why would you bother with such little diversification?

[ILikeDividends]

Why do people still buy the S&P? Schwab has SCHB for 1 bp more. Thousands of stocks vs 500 large cap only. Why would you bother with such little diversification?

If your asset allocation has different goals for different allocations, you can achieve a more targeted allocation and re-balance as often as you like with no transaction costs using these:
 
SCHE - Emerging Market
SCHF - Foreign Developed Market
SCHX - Large Cap Domestic (i.e., S&P 500)
SCHA - Small Cap Domestic (i.e., Russel 2000)

Essentially, you can get even more diversification using the above 4 ETFs than you can get with SCHB, and maintain better control over the allocations.

I don't like using an ETF for the fixed income portion of my AA, but if you don't care about that, Schwab has a no-fee ETF for fixed income too.