Most diversified fund managers claim to be bottom-up in their approach. Funds use roughly similar methods of valuation
01-Dec-2006 •Devangshu Datta
Top down or bottom-up? Investors spend a lot of time debating that question. Some say that you should have a narrow focus on the best companies, ignoring broader considerations of sector, macro economic factors, etc. Others say the sector is more important than the specific pick. In a broader sense, the same debate is held about asset allocation. Some say that you must pick exactly the right instruments, others that you must pick the right class of instruments.
In a global sense, this argument also rages around nation states. Do you look for a booming economy (India or China) and then focus on the best businesses in those places? Or do you pick the best global businesses wherever these are headquartered? There are logical arguments for and against either style. When the macro factors are good, an entire range of instruments, or every company in a sector, may give great returns. On the other hand, "desert flowers"- businesses which consistently perform even in adverse circumstances, often offer something special in the way of management skills or competitive advantage.
The investment style may be different but it often leads to the same portfolio picks. A top-down investor looks for a sector, which is in the sweet spot in terms of business cycle and favourable macro-factors. Then, he looks for companies operating inside that sector and tries to make comparative judgments.
A bottom-up investor ignores sector dynamics and concentrates on company financials. But he may end up short-listing several companies from the same sector because these are all doing well due to the same favourable factors.
In practical terms, top-down may be a more efficient process. There are fewer sectors than stocks; that makes it easier to make the first cut. And, the initial choice is always the most difficult part of an investment selection process.
One interesting top-down method could be to look at mutual fund holdings across different sectors. Most diversified fund managers claim to be bottom-up in their approach but that really doesn't matter. Funds use roughly similar methods of valuation and selection. And, we can aggregate their holdings and classify these across industry sectors as has been done in the attached table, where 14 sector divisions have been used.
This chart gives us a top-down view of what may have been a "bottom-up" selection process. But the macro-factors are clearly favourable for the sectors where equity funds are overweight in aggregate. And, they are clearly unfavourable for the sectors where funds are underweight. They may be changing for either better or worse in sectors where the funds are showing a clear shift in preference.
We can look at this data in another way. A sector with "average representation" in aggregate fund holdings should have roughly 7 per cent of corpus allocation. If the representation is significantly more (or less), there is a bias in favour of (against) the sector in question. Obviously this is a very crude rule of thumb since different sectors have different aggregate revenues and profitability. But we could use it as a first cut.
Over the past year, on a month-by-month basis, funds have been consistently overweight in technology, engineering and financial services. They have been noticeably underweight in textiles and consumer durables (CD). The CD aversion is understandable on the grounds that most industry majors are unlisted. In textiles, there is apparent consensus that the macro-factors are unfavourable. In most other sectors, holdings are fairly close to 7 per cent and haven't shifted too much.
Obviously tech, engineering and financial services are "hot" with over 35 per cent of total corpus allocated between them. A potentially hot sector is construction - several big stocks have listed here in the recent past and the infrastructure/ real-estate boom has started to translate into market valuations. Fund holdings have risen from 6.6 per cent a year ago to nearly 9 per cent by October. Once Parasvnath (listed in late November) and DLF (where an IPO is awaited) come through, construction allocations will almost certainly rise again.
Energy is another likely "hot" sector on the basis of rising allocations and, unlike construction, the universe of stocks hasn't changed much in energy. Are we stretching the logic if we suggest that Cairn Energy will do well on the basis of the rise in energy sector allocations?
If these trends hold, we will see a concentration of about 55 per cent of total corpus within five sectors fairly soon. That implies that the business cycle is narrowing. Even if funds employ bottom-up stock-picking methods, their allocations are becoming increasingly sector-specific.