Time to get defensive?

The Footsie hit a low in March 2009, at about 3530. Two years later, it stands at about 5957; an increase of about 70%. That's a huge rebound, and investors may be wondering where the market is headed next. What areas should we look at to maximise our returns: cyclicals, growth stocks, defensives, bonds, something else maybe?

In his article Focus: Searching for Growth, Chris White argued:

The greatest value across all asset classes at the present time lies in large-cap, blue-chip, high-yielding equities. Equity income funds have come under fire for all being invested in the same stocks in this area of the market ... with UK FTSE 100 companies receiving 70 per cent of their earnings from outside the UK, our globalised stockmarket is not just aligned to UK GDP, but more to a blend of international economies, most of which can be expected to grow faster than the UK in 2011 ... Investors should start to look to sectors which benefit from rising inflation, interest rates and bond yields such as insurance, oil, tobacco, utilities and food retail.

The Motley Fool quotes the following recent statement by Neil Woodford in this article:

I am invested in some of the best quality companies in the UK stock market on ludicrously cheap valuations. In my opinion, this is the best investment opportunity since the tech bubble in 2000.

As an experiment, I sought confirming/disconfirming evidence for the bullish case on defensive shares. I chose 7 large-cap defensive companies at random: £AZN, £BATS, £CNA, £DGE, £TEP, £TSCO, £ULVR. I compiled the following table:
       PER PILE Y10  Y0
AZN    6.8   10 1.4 5.7
BATS  13.6   70 5.6 3.5
CNA   12.6   50 1.2 4.5
DGE   15.0   50 3.1 3.4
TEP   15.7   50 1.6 4.9
TSCO  11.7    0 1.9 3.8
ULVR  14.4   40 2.6 3.8
The key to understanding the table is as follows:
PER - is the latest rolling Price/Earnings ratio
PILE - "percentile" - the percentile to which a company's current PE ratio fits over the last decade. So, with AZN on 10, it means that only 10% of the time has its PER been lower. PERs have been taken at approximate year-end snapshots, rather than being calculated on a daily basis. They should be considered as indicative, rather than precise.
Y10 - the yield on the company a decade ago, in percentage terms
Y0 - the latest available yield on the company, also in percentage terms

Critics might argue that my sample size is small, and I accept such charges unreservedly. We notice two things:

1. the median PILE in the sample above is 50, indicating that, on the whole, there were as many chances as not to buy the companies at more attractive PERs. £AZN and £TSCO stand out as two companies that are on historically cheap terms over the decade, suggesting that there may be significant opportunities for selective bargain-hunting.
2. Defensives are now much more attractively valued than they were a decade ago - as evidenced by the fact that the yields have increased (Y10 compared with Y0) in all cases except £BATS. This shouldn't come as a major surprise to most people, though, as markets were at record highs at the turn of the millenium. Interest rates have declined over the decade, whilst yields on defensives have gone up.

In conclusion, it appears that defensives are cheaper than they were a decade ago, although it would be wrong to say that they are cheaper than they've ever been for the last decade.  That doesn't quite square with Woodford's statement about them being the cheapest they've ever been. Far be it for me though, in genuine humility, to argue with a man of Neil's stature. Perhaps he is arguing that the yield gap between general interest rates and top-quality inflation-hedging companies is very bullish for the latter.