Benchmarking portfolio performance 4 comments

(P.S: Sorry for any disturbances the advertisements above may have caused you)
Benchmarking is the process of comparing one's operating performance with the industry average and top performers in order to gauge one's progress. This is often one of the best reality checks; numbers don't lie, especially when accurate and highly context-relevant indicators are employed and honesty is applied to the self-assessment process.

Those who buy unit trusts would be familiar with such benchmarking with respect to the fund management industry. The relevant MSCI Index is usually the benchmark being employed; funds investing in specific countries would benchmark against the relevant MSCI country index, those specialising in regions (eg. Asia-Pacific) would employ the regional index for comparison (MSCI Far-east in our example).

If the professionals do it, surely it is a tool worth exploring for the small investor. It is hardly time-consuming; the STI and the SESDAQ indices are reported in the papers every day and are highly relevant for the local stock market investor. I find that such benchmarking gives me a good idea of what drives my portfolio growth: market, sector, or company (those reading my earlier blogs would know I believe these three are equally important: see Stock Selection Part 1).

The thing to note about benchmarking is that one can't be too short-term about it, in the way that one benchmarks against the index every week or even month. Typically a quarter, or three months, is the shortest period to warrant benchmarking performance against. This, of course, extrapolates from the common sense that one is dissipating energy unjustifiably through daily stock monitoring. The second thing is that unlike normal industry benchmarking, where one benchmarks against the best in the business, it is often silly to apply this principle to portfolio management. That's because rotational sector and theme plays vary all the time and it is impossible to anticipate their timing; one would tear his hair out if he continually compared the relative performance of his stocks with that of the top gainers and fretted about how much money he had "lost" by not buying into the latter.

As much as overly-aggressive benchmarking should not be employed, it also doesn't make sense to adhere blindly to one single index all the time. In late 2003, the SESDAQ outperformed the STI, doubling itself while the latter rose ~20-30%. The trend reversed in 2004-05 when the STI continued to rise while the SESDAQ tanked. The investor who had adjusted his asset allocation well would have captured good gains in both periods. The investor who has been invested in small caps all the way might say: look, at least I'm market-performing in comparison to the SESDAQ through 2004-05; but what's the point? That's like a poor man in a developed country comparing his salary to someone in a Third World country; it's good for self-consolation but does nothing much for the drive towards self-improvement.

The above point might seem paradoxical with the earlier view that one should not continually compare portfolio performance with best performers, but my main point is this: through a flexible system of benchmarking (ie. different comparison indices used at different times), one improves his understanding of market valuation cycles, improves his asset allocation strategy such that he will be better prepared for the next one, and in the meantime is constantly driven to adopt a flexible mindset with regard to buying/selling. The small investor should utilise fully his advantage of being small: he can enter into and exit positions easily (hence adjusting his asset allocation, say from fully-valued small caps to undervalued blue chips in early 2004) without any market impact; the institutional investors (eg. fund managers) cannot do this easily (another reason is that their fund charter only permits a certain investment category) and that's why they often peg performance benchmarks to one single index.