Properly Benchmarking Your Portfolio's Performance

by: Roger F. Goodrich

Benchmark your portfolio

Did you ever wonder if your portfolio was keeping up with the market? There is an easy way to find out. Let's use the S&P500 as the standard, or benchmark. Simply divide the value of your portfolio by the benchmark, do this periodically and track the results. Assume your current portfolio value is $400,000 and the S&P500 is 1350. The quotient is 400000/1350 = 296.3. Next time, the portfolio value has increased 3500 and the S&P 500 is up 10. The new values are 403500/1360 = 296.7, so you have gained a little. [Side note: I do this calculation weekly, I have my portfolio segments in an on-line portfolio site such as Morningstar or I could also use my online brokerage site. The value of the S&P500 is obtained using symbols ^GSPC, $SPX or even the value of ETF SPDR S&P500 (NYSEARCA:SPY).].

It is more meaningful to perform one more operation in this calculation, to gain more visibility in changes as well as to account for additions to and/or withdrawals from the portfolio. This is accomplished by subtracting a number close to the tracking number, say 300 in this example. This would yield 296.7 - 300 = -3.3. The exact value of this initial number (call it number N) is not critical, only so the initial result is near zero. Each time period subtract this number from the quotient. Changes to the portfolio, buying/selling shares, is handled by adding (for putting money in) or subtracting (for taking money out) to the number N that amount divided by the S&P500 value.

For example: add $5000 to the portfolio (from a 401k contribution). The new number N to be used to subtract from the quotient is 300 + 5000/1360 = 303.7. The new portfolio is larger but not due to market variations, so the number N has been increased to compensate. This allows us to track only the changes due to market conditions.

That is all there is to it, record the numbers on a periodic basis, input them in a spreadsheet and track how you do over time. You will more than likely find, as I did, that there are short term and long term variations in the result. The figure below shows my portfolio from 1 Oct 2010 to 3 Aug 2012, in red vs. the S&P500 (re-scaled) in green.

This is a retirement portfolio, low beta. When the market goes up; the portfolio goes up, but not as much -negative in a relative sense. When the market goes down, the portfolio goes down, but not as much - positive relatively. During the time period shown, the S&P500 was flat to slightly up. My portfolio was flat to slightly down. Just what you would expect from a lower beta portfolio. My conclusion: top level, OK. A measure of the relative variability of the S&P500 and the portfolio can be found by dividing standard deviation by the average for each. The results are: S&P500, 5.6%; Portfolio, 3.5%.

I have split my portfolio into 5 segments: 1) Dividend Growth (30%), low yield/high dividend growth stocks, such as Procter & Gamble (NYSE:PG), Illinois Tool Works (NYSE:ITW), McDonald's (NYSE:MCD), Owens & Minor (NYSE:OMI), Norfolk Southern (NYSE:NSC), Darden Restaurants (NYSE:DRI), Hasbro (NASDAQ:HAS). 2) High Income (30%), high yield/lower dividend growth, such as Omega Healthcare Investors (NYSE:OHI), Enterprise Products Partners (NYSE:EPD), Shaw Communications (NYSE:SJR), Avista (NYSE:AVA). 3) Core (15%), ETFs with some yield/dividend growth, such as Vanguard FTSE All-World ex-US (NYSEARCA:VEU), Vanguard Industrials (NYSEARCA:VIS), ishares S&P Global Infrastructure (NYSEARCA:IGF), Wisdom Tree Commodity Country Equity (NYSEARCA:CCXE). 4) Mélange (15%), some less pristine examples of the above plus stocks with high beta (for up markets), such as Triangle Capital (TCAP), Annaly Capital Management (NYSE:NLY), Vanguard Small Cap (NYSEARCA:VB), Microchip Technology (NASDAQ:MCHP). 5) Bonds (10%), such as SPDR Barclay's Capital High Yield (NYSEARCA:JNK), Templeton Global Income (NYSE:GIM).

One reason to split the portfolio into segments is that it is intended to drawdown from these segments (at different rates) as a consequence of the retirement portfolio withdrawal scheme detailed in the 3 part series; Life Cycle of Retirement Portfolios, Part 1, Part 2, Part 3.

The graph below shows how this procedure works on 4 of my portfolio segments. I use this approach, as one way, to keep tabs on price performance of the portfolio. See my articles on dividend growth for how I track the income metrics.

It is clear that the Dividend Growth and High Income segments are well behaved with respect to keeping up with the market. Core, the green curve, was OK up to the middle of the time period. This segment is heavily weighted to international ETFs, which haven't performed well lately. Mélange is a mix which obviously hasn't done well vs. the S&P500. Part of the reason these last 2 segments have under-achieved is that they contain some higher beta equities which don't always perform well in a sideways market, which we have had. The takeaway for me here is to watch these segments. If they perform well in a strong up market, then OK. If not, then there is cause to cull under-performers. At the moment, no single component stands out as a candidate for selling.

Keep in mind that it is the changes in these curves that matter. If a line is flat for a portion of the time period, no matter where it lies on the chart; then during that time, that portfolio segment is keeping pace with the market.

The situation becomes clearer if we combine segments: High Income + Dividend Growth and Core + Mélange. This is shown in the graph below. The time period is shorter, from 3 Jun 2011 to 3 Aug 2012.

Core and Mélange together make up 30% of my portfolio. They both contribute dividends and have some measure of dividend growth. They provide capital gains potential. They provide diversification. Their time will come. It will, won't it?

Disclosure: I am long PG.

Additional disclosure: I am long all stocks and funds listed in the article except SPY.