Thursday, September 24, 2009

Steroid Soup By Tim Bond For Valuation Pros

FT Alphaville features the latest note by Tim Bond, the head of global asset allocation of Barclays Capital.

I do not focus on the environment that "is ripe for the inflation of asset bubbles" this time. I go for equity market valuations - earnings and PE (price to earnings) ratios in particular.

My eyes were paralyzed by the following chart as valuations appear decent, click to enlarge, courtesy of Barclays Capital.

I have no idea about the "equilibrium PE model" by Barclays that, by the way, serves well for the justification of bubbly valuation in Internet frenzy. However, the path towards lower PE ratios in the next 5 years feels intuitively right.

The problem is that Tim Bond is not specifying his US market, but PE ratios look decent.
Well, let's look at S&P500, where we have access to historical earnings and PE data. Here are S&P 500 Earnings (Operating, As Reported & Core) and Estimate Report, includes Divisors and Aggregates, Core Breakdown, and Dividends. And here are S&P500 historical average price to earnings ratio.

As You open the S&P500 Earnings sheet from Standard & Poor's, the puzzle starts. You get operating, as reported and core earnings. Which one to use? Well, to make the decision, one should have a basic idea of different measures. S&P has a note with explanations here, see Table 2 on page 4 for the main differences between Operating and As Reported earnings.


Credit strategists at BNP Paribas have been wondering:

Over the past equity bubble decade, it has become fashionable for equity analysts to concentrate on Operating earnings as opposed to As Reported earnings, which factor in write-offs and restructuring charges (Charts 1 and 2).

While the difference between the two measures was insignificant until the internet bubble, that difference has grown significantly to the extent that operating earnings look like numbers plucked out of thin air with little resemblance to economic reality. As credit analysts, we are taught that, for a given revenue base, rising costs lower profits, raise leverage and lower creditworthiness. How equity analysts can ignore this fundamental credit analysis is unfathomable to us.

Click on charts, courtesy of BNP Paribas.



Indeed, it looks like Tim Bond is using Operating Earnings for his show, but S&P and credit strategists at BNP Paribas suggest that As Reported earnings are more appropriate. Probably the ignorance of fundamental analysis has been driving us into credit crisis, and still seem to be dominating the equity stage?

Historical data from S&P since 1936 show that average S&P 500 Historical As Reported PE Ratio based on trailing 12-month earnings is almost 16, but the median is 15.5.

12-month forward looking earnings estimates are varying, but let's look at As Reported estimates, that are done by top-down strategists (interestingly, bottom-up estimates are only for operating earnings). 12-month forward estimate is 41.49 USD EPS (earnings per share), which at 1050.78 index closing on my screen tonight makes PE ratio of more than 25. The EPS estimates for full calendar year of 2010 stand at 45.84 USD today, what at current index value makes PE ratio of almost 23.

Well, let' s go bullish to the peak cyclical As Reported earnings in the 2nd quarter of 2007 at 21.88 USD in the quarter, which simply annualizing makes 87.52 USD EPS annualized. At historical average PE of 16 it makes index value of 1400. At the minimum PE of 5.9 recorded in 2nd quarter of 1949 the value would be 516.

In the 2nd quarter of 2009 the As Reported earning were 13.5 USD EPS in the quarter, which simply annualized arrives to the yearly EPS of 54 USD, multiplied by average historical PE of 16 is 864.

For the last 20 years the quarterly As Reported "trend" earnings adjusted for cyclicality are somewhere around 14 to 15 USD, which makes something like 56 to 60 USD of yearly EPS. Multiplied by historical average PE it makes index value of 896 to 960.

One should guess whether massive fiscal stimulus and aggressive monetary policy is justifying above or below average PE ratio. What is the margin of safety or premium to endless next cycle? You decide!

As my old friend says - how many methods of valuation do you know?

2 comments:

  1. Interesting article...

    The problem comes with trying to use as reported earnings at the end of an earnings cycle when earnings have just been obliterated. If we are going to use GAAP earnings, probably better to do some smoothing and try to find the overall trend.

    Overall long term earnings trend is fairly stable, and earnings tend to grow at above average rates after getting clobbered during a recesssion, until they get back to that long term trend... So with your smoothed GAAP earnings, you might want to include even some estimates going forward for the next 6-12 months...

    The DJ PE went infinite near the bottom in 1932, the best time to buy stocks in the 20th century.

    ReplyDelete
  2. Hi Anonymous,
    Ben Graham also used "As Reported" earnings approach, and was smoothing out as a 10 year moving average ... according to that approach the yearly earnings stand at 48.78 USD, and the PE (Graham and Dodd PE) was used as 16, this means something like 780 for S&P500...

    Another approach would be using data from Robert Shiller, and, according to his "real earnings approach", the 10 year moving average is at 55.90 USD ... interestingly, his average PE ratio for 10 year average earnings since 1870ties stands at 16.34 ... if using PE ratio of 16, then the index value of S&P500 comes at 894 ...

    Quite obviously, market is about priced for the sweet spot.

    ReplyDelete