SMP = Net income less ((Other income and R&D expenses) * (1 - tax savings resulting from these expenses)
The margin value is calculated as a percent of revenue.
Per share values should use visible diluted shares as the divisor.
Other income is non-recurring exceptional items.
R&D is presumed to be a worthwhile investment that will increase future revenue and income.
Sales and Admin expenses tend to grow uncontrollably with revenue and this metric correctly encapsulates the level of control managment is exercising regarding these expenses.
Taxes are significantly increased by stock option compensation schemes, and the ratio properly reflects the extra costs imposed on shareholders by such schemes.
The ratio validates to external investors the exact return from R&D that would be needed to justify that R&D.
For growth companies, the per share value is a better metric than earnings. If you can imagine a "terminal" annual sales value, then the margin ratio determines what those sales would contribute to shareholders if management has some disposition to paying shareholders
Share price as a multiple of SMP margin
Because SMP is a better metric than EPS, there is a more universal multiple that can be used to calculate fair share price, though there is still substantial company-specific modifiers.
As a starting point, use a 5% fair interest return for a 20 year period in the company's weighted markets. While national rates may be lower than this, they reflect supply disruptions and forced purchases that may be temporary, and a real AAA credit rating, would have reasonable investors accept such returns in free conditions.
The most universal adjustment is the disruption risk factors to a company. Crocs, Sony Walkmans, Blackberry's and Myspace all had spectacular sales growth at one point. Many companies do not have visible sustainability challenges, though looking backwards they are obvious, and caused price overvaluations.
For simplicity assign an optimistic terminal sales value that seems to have a 50% chance of occuring. If they are disrupted, they are most likely to churn their wheels breaking even for a while. Crocs, Blackberry, Sony, Myspace all still exist. Their market price is substantially lower than their peaks but there can be hope that they will rebound. Using a 50% likely sales target with the breakeven fallback position means that a shareholder should expect a 10% return at the sales target. There is still substantial risk for overvaluation using this technique, but disruption doesn't happen until it happens, and obtaining a fair price based on the likelihood of disruption is still appropriate. Buying or not becomes a decision based on the fear of disruption.
Of the 4 companies mentioned, Sony is the only one that has paid a dividend. Since 1990, it has paid about $10, and its shares are up $3. So dividend payout and management predisposition for payout is critical to mitigating any risk of disruption. Some companies (MCD) are likely to pay (close to) their entire market value in dividends over the next 20 years, and so its shareholders are substantially insulated from disruption. Other companies have dual class share structures that make their board unaccountable to anything but bare minimum Delaware regulations. They will not pay dividends, and will not accept reasonable takeover offers as their piggy bank can better continue to serve them if they keep control of it.
The final adjustment to the SMP multiple is incorporating this management disposition and accountability factor into an adjustment to the previous value. A company expected to pay 0 dividends over the next 20 years should be worth 75% of its disruption discounted value. A company expecting to pay 100% of its market value in dividends over the next 20 years should be worth 100% of its disruption discounted value.
With respect to takeover potential, as important a factor as management predisposition, is the market size of the firm. A company worth $2B still has enough potential fishes large enough to pay $4B for it. $20B and $30B companies, not so much. While there is substantial overlap between 0 dividend paying companies and those uninviting of takeovers, there is a qualitative adjustment combined with the previous dividend discount that should bring companies as low as 50% of their disruption discounted value based on their size and level of board control and shareholder disdain.
Blackberry was once a $200+ stock. Never paid a dividend. At the time they were too big to be taken over. Today, much of their remaining value is based on an actual eagerness to receive a reasonable takeover bid (~$15/share?). Even if you would evaluate a low sustainable disruption discounted SMP company value, there would be a premium for takeover potential.
Some company valuations (with 35% tax benefit of R&D expenses)
To get to a $30B value, you need to assume $27B in sales and a return to 2013 SMP margins of 23%, and the growth deterioration at extreme expense expansion doesn't support this.
In my view, LNKD has been facing disruptions to the market opportunities it saw in 2013. Its reaction to those disruptions has been overspending and the consequences of those reactions are described by the lower SMP margins.
Twitter has worst problems in that it would still be unprofitable if it had no R&D expenses. Its not the case that young companies simply have low ratios, as its 2013 SMP margin was a low but respectable 18.5%. What its done since simply hasn't worked.
While there is some deterioration over last year, at $27B terminal sales, the disruption discounted value would be $90B. A drop of $10B from last quarter and last year.
Still its R&D spend gives it the power to stiffle LNKD and TWTR's growth.
If FB will spend $4B per year in R&D, it needs $12B in cummulative incremental sales to justify it. If those sales can be amortized over 5 years, then it needs a $2.4B annual sales increase to just break even on the R&D. It is meeting this threshold so far.
LNKD on the other hand at its current SMP margin, needs to turn $700M in annual R&D into $7B in incremental cummulative sales. It is not expected to meet the break even $1.4B amortized annual sales increase, and so either its R&D is wasteful or the sales and admin costs are too expensive to obtain its sales.
Google improved its SMP margin over last year. At $150B expected sales 20 years from now, there is a $450B valuation. A 50% discount factor is still appropriate though there is a small unlikely probability of divdidend as a result of the ageing and potnetial retirement of some of the founders. At least more likely in the following 20 years after the first 20.
At $225B fair value, GOOG is the least overpriced of the companies mentioned, though it is not at all immune from disruption.