Amazon ($AMZN) recently released their quarterly numbers with a 38% drop in Q2 profit. This did not concern the owners nor the management of $AMZN.
$AMZN is an extreme case of the need for an economic accounting framework when analysing investment opportunities. The owners and management of $$AMZN are not margin driven (in the short term), ROE aware (in the short term) and would not consider their cost of equity over time.
What makes us different is this: We are genuinely customer centric, we are genuinely long term oriented and we genuinely want to invent. Most companies are not those things: they are focused on the competitor rather than the customer. They want to work on things that pay dividends on two or three years, and if they don’t they move on to something else. They prefer to be close followers rather than inventors because it is safer. – Jeff Bezos
How do you construct a robust discounted cash flow, sum of the parts or peer relative $AMZN valuation?
We can attempt a sum of the parts valuation for $GOOGL. The division of the $GOOGL’s assets is sufficiently discrete. For $AMZN this is problematic as only AWS could be stripped out and valued separately. This is possible because the AWS customer is different to the Amazon customer. Amazon is a customer centric business. Attempts to value its business divisions separately runs counter to the way Amazon is managed.
Peer comparison is possible but the consequences will make most investors pause. We can place a value on $GOOGL’s search business, conservatively this is in the order of US$400 billion. Consider $AMZN in the US as a search business.
L2’s Scott Galloway believes that AMZN can access 70-80% of US disposable income within 20 minutes .
Think of the resources devoted to this grail by other entities. $AMZN is already there.
Clinically we can view $AMZN as superior search business to $GOOGL. We may also conclude that $AMZN’s current market cap only values $AMZN as a search business. Everything else is free!
If I were to put this to an investment committee I suspect I would be demoted to junior Materials sector analyst. Clearly it is time for some beer coaster investment analysis.
For the past month I have not been able to find an investor that thinks Amazon is wildly overvalued at current prices. If Amazon’s current market cap equals its current valuation then we can derive an economic profit figure by utilising a sensible earnings multiple. If we apply a multiple of 25X we arrive at an economic profit of US$19.5 billion. A long way from the US$2.4 billion statutory profit. Conveniently it is very close to Google’s 2016 accounting profit and about 43% above Walmart (Walmart’s current market cap is US$244 billion).
$AMZN’s annual revenue to December 2016 was US$136 billion. Our economic profit of US$19.5 billion represents an NPAT margin of 14%.
$AMZN’s economic profit of US$19.5 billion can withstand some scrutiny. Obviously it would take more than the previous paragraph to get $AMZN past an investment committee. I am not an $AMZN evangelist, I use as a justification for developing a robust economic accounting framework.
Whilst we all recognise that statutory accounting and economic accounting serve different users, it is an area of concern when statutory accounting principles cloud a company’s economic profit.
Why do too many financial models have maintenance capex equal to depreciation?
Why do too many financial models assume that capitalised R&D is a suitable valuation for a company’s R&D?
Every fund manager has a set of economic accounting principles, it is the investment process section of their prospectus. In its most basic form, a proxy for a company’s economic profit is the company’s free cash flow.
As investors we tend to focus on an economic P&L. Rarely do we attempt to establish an economic balance sheet (adding one massive goodwill entry doesn’t count). If you are struggling to find a justification for establishing a company’s economic balance sheet so am I. Despite this I am always on the lookout for an asset that exhibits linear deterioration. Surely the first asset depreciation table was based on observation and not mathematical convenience.
There are a number of balance sheet items that require an economic accounting framework. Capitalised expenses and contractual commitments immediately come to mind. How do software companies get away with capitalising R&D when the R&D results in a product that has no pricing power with its clients?
If any company commits 40% of revenue to R&D it demands attention (rather than accepting software company mantra). How can 40% or 50% of R&D spend be capitalised if it provides no lasting economic advantage to the company?
Software companies tend to exhibit high statutory accounting operating margins but very average economic accounting margins.
I am not sure of a solution but I am convinced that there is a wealth transfer going on here. Not all software products/services are critical but some provide the infrastructure (enterprise application software) necessary for the day to day functioning of an entity. Why do these companies sell their products for less than they are worth?
Discounting a company’s future cash flow is not easy without a time machine. Yet we tend to develop great confidence after constructing a ten year model and a bit of scenario analysis.
On the face of it, establishing $AMZN’s economic profit is flaky. However, it would have less flaws than a ten year financial model.