February 20, 2003; Comments Behind A Slide on Valuation
presentation on Energy Trusts and Partnerships at the Peters Energy
Conference in Lake Louise, Alberta, Canada, summarizes many of the points on
those income stocks we have made in Meter Reader, Natural Gas Royalty
Trusts and Independent Stock Ideas.
More than twenty thousand investors have apparently hit the page on our
website that reproduces the slides. Because
valuation analysis underlies all of our conclusions we add some discussion of a
slide on valuation that prompted questions.
The questions referred to the equation below that appeared on a slide
entitled “Valuation - McDep Ratio”.
for one year is fairly straightforward for energy companies, except for
potential accounting deception more likely to occur with high debt, high greed
issues. The multiple is more
complicated. In the case of our
weekly calculations for royalty trusts we try to quantify most of the variables
that affect PV. When PV is
separated into Ebitda times PV/Ebitda we find that Ebitda for the first year is
influenced most by one-year commodity futures.
It turns out that PV/Ebitda is influenced mostly by reserve life
initially and then by ongoing fluctuations in six-year futures prices that can
be different for oil vs. gas. The
latter point can be a differentiator of oil vs. gas when the ratio of six-year
futures to one-year futures for oil is a lot different than for gas.
After doing a lot of present value calculations one gets a sense for PV
as a multiple of Ebitda.
pipeline bulls in effect say the Ebitda will last forever.
The oil and gas bulls in effect say that prices will be higher to offset
volume decline, or that volume will not really decline as much as one might
valuation tables are set up for users to judge our estimate of PV.
One would first consider whether our Ebitda estimate is up-to-date.
Some estimates are not as fresh as others.
Then one can look at PV/Ebitda in our - 2 (e.g. Tables 1-2, 2-2, 3-2)
valuation tables and ask oneself whether differences in the mix of business
among stocks justifies differences in PV/Ebitda.
could say that most investors judge stocks on a P/E basis.
We are not comfortable doing that in energy for two reasons.
First, debt is too important in many cases not to be handled explicitly.
That is why we use the McDep Ratio because it is unlevered. We cited work in the Financial Analysts Journal by Martin
Leibowitz, which makes the argument that P/E ratio, by definition a levered
ratio, should be adjusted for debt. In
other words, high debt companies can look too good on the P/E ratio.
One might even extend that as an argument that current P/E ratios
overstate value in the market as a whole because most companies have much more
debt than historically. In energy, infrastructure companies have gorged on debt, but
not integrated companies and producers for the most part.
we don't think earnings are meaningful for energy valuation because, as the
companies are capital intensive, earnings depend too much on historic costs,
which by definition will be low because of inflation for mature companies and
high for new companies. It is
better to have earnings than not, but we wouldn't pay for it.
Ebitda estimates are theoretically independent of historic costs.
Ebitda has its own drawbacks that are overcome partly by our use of next
twelve months estimates that can be adjusted for other distortions in past
results that we may identify.
Finally another advantage of PV/Ebitda, or EV/Ebitda, is that it is easy to relate it to acquisitions of businesses as well as to the pricing of stocks. And that leads us to intriguing questions as to why a particular business might be priced so much higher or lower in the stock market.