Benchmarks
Four indices to measure your portfolio against, and why the choice matters more than it looks. Beta, alpha and the capture ratios are all relative to whichever one you picked.
On this page
A benchmark is the alternative you are being compared against: what would have happened if you had simply bought the index instead.
The choice is not cosmetic. Beta, alpha, tracking error, the information ratio and the capture ratios are all defined relative to it. Change the benchmark and every one of those numbers changes.
The four available
S&P 500. The 500 largest US companies, and the standard US large-cap reference.
MSCI World. Around 1,500 large and mid-cap companies across 23 developed markets. The broadest of the four, and the closest to "global equities" as a default.
EURO STOXX 50. The 50 largest blue-chip companies in the eurozone. Narrow and concentrated.
STOXX Europe 600. 600 large, mid and small-cap companies across Europe, including outside the eurozone. The broad European option.
Choosing one
Pick the one that represents what you would have done otherwise. That framing settles most of the difficulty.
If your alternative to picking stocks is a world tracker, MSCI World is your benchmark, whatever your portfolio actually contains. If you deliberately invest in European companies, STOXX Europe 600 tells you whether the selection added anything over the region as a whole.
What does not work is choosing the index your portfolio most resembles in order to look good against it. A European portfolio measured against the S&P 500 in a strong dollar year is being compared with a currency move rather than with an investment decision.
Currency is inside the comparison
Everything in Gylder is presented in euro, including the benchmark.
So a US index measured from a euro perspective carries the dollar-euro move inside it. In a year where the S&P 500 rose and the dollar fell, the euro-denominated benchmark rose by less, and your performance against it improves without anything about your holdings changing.
This is not a distortion: it is the honest comparison, because holding a US index as a euro investor would have exposed you to exactly that currency move. It is worth knowing before attributing the result to selection.
Why it might show nothing
Benchmark-relative measures need benchmark data covering your whole window. When it is missing, the whole group shows a dash rather than a number, which usually means the price cache for that index is not warm yet rather than anything being wrong with your portfolio.
What a benchmark cannot tell you
Whether the comparison is fair. A concentrated portfolio of five holdings and a 1,500-company index are different propositions, and beating the index with five names says as much about risk taken as skill applied.
Whether you should have done something else. Gylder does not recommend a benchmark, and does not score you against one it picked. The default is a starting point, not an opinion about your strategy.
Anything about the period you did not select. Benchmark-relative figures recalculate with the window, and a portfolio that beats an index over five years can trail it over one.
Was this helpful?