There have been buzzes about Google's (GOOG) stock price, which exceeded $1000 last Friday after its Q3 earnings release. Can the $1000 price be justified by the fundamentals? We look at Google's free cash flow valuation and infer the set of assumptions the market might be using to price Google shares.
The free cash flow valuation approach equates Google's enterprise value to the present value of its future cash flows. Key to this approach is the forecasts of the three components of cash flows: those from operations, from investments, and from changes in net working capital. Google's net working capital is miniscule relative to its operational size; and its valuation is more sensitive to the forecasts on the operating and investment cash flows. Three important assumptions go into the forecasts: 1) growth rate, 2) profitability, and 3) rate of investments on long-term assets. Related to these metrics, below are Google's numbers in the past several years:
Sales growth rate
(Source: Google Finance)
The sales growth rate for 2013Q3 is relative to 2012Q3. The 2013Q3 numbers for EBITDA/Sales and for Long-term assets/Sales are based on trailing 4 quarters. (Long-term assets = Total assets - Current assets.)
Although sales growth rate for 2013Q3 is low relative to the past three years, analysts' consensus revenue forecasts for 2013 and 2014 are $59.62B and $69.32B according to Yahoo Finance, translating into annual growth rates of 18.82% and 16.27%. The EBITDA-based profit margin displays a visible downward trend. At 2013Q3, although the trailing-4Q margin is only slightly below that for 2012, the number for Q3 alone is below 30% (29.83%). Long-term assets as a ratio to sales revenue jumps up from 50% to 66% in 2012, reflecting the impact of the Motorola Mobility acquisition.
Other financial ratios relevant for forecasting are relatively stable over the past years (trailing 4Q number for 2013Q3):
Statement tax rate
Net working capital/Sales
(Source: Google Finance)
A lower tax rate for 2013Q3 is the result of loss harvesting from the Motorola Mobility division.
The following assumptions mechanically follow analysts' consensus forecasts and past financial ratios. They turn out to support the current market valuation of Google.
1) Take the consensus revenue forecasts for 2013 and 2014 ($59.62B and $69.32B). For 2015 to 2017, take the consensus 5-year long-term growth rate forecast of 16.40%, which is close to the consensus revenue growth rate forecast of 16.27% for 2014. (These forecasts are from Yahoo Finance. 16.40% is actually the 5-year growth rate projection for EPS; however under the constant-margin assumption it would be equivalent to the revenue growth rate).
2) Assume a steady state in 2022, 10 years from now. The growth rate in the steady state is 3.5%, which is a mix of the long-term domestic and international GDP growth rates. Between 2017 and 2022, assume that the revenue growth rate linearly converges from 16.40% to 3.5%. FCFs after 2022 are assumed to be growing perpetuity at a growth rate of 3.5%.
3) Assume a constant EBITDA/Sales margin of 31.84% (the 2013Q3 number) for all future years.
4) Assume a constant ratio of Long-term assets/Sales, at the 2013Q3 number of 64.45%. Long-term assets are forecasted based on this ratio and the revenue forecasts. Investment cash flows are then inferred from long-term assets changes and depreciations.
5) The depreciation rate and the Net working capital/Sales ratio are all kept at the 2013Q3 level, i.e., 10.65% and 1.55% respectively.
6) Finally, assume an effective tax rate of 20%, similar to the statement tax rate in the past years.
Below are the baseline forecasts under these assumptions (in $billions):
You can find the details in a spreadsheet file here.
In addition, the following assumptions are made in order to determine the discount rates for calculating the present values of FCFs. Stock beta is 1.0 (0.97 on Yahoo Finance). Riskfree rate is expected to gradually increase from 3% in 2013 to 4% in 2018 and stays at that level afterwards. Equity premium: 6% (close to the average equity premium in the past 80 years). Cost of debt: 6%. Ratio of debt/firm value: 5%. (Since Google has little debt, the cost of debt and capital structure assumptions are not material.) These assumptions translate into an annual WACC of 8.79% in 2013 and a gradual increase to 9.74% by 2018.
Here is the nutshell of what comes out of the baseline model. The present value of future FCFs sums up to $210B. Adjusting for cash holding of $57B and debt of $10B, Google's intrinsic equity value amounts to $343B. With 339.24 million shares outstanding, the intrinsic stock price estimate is $1010 per share, pretty close to the market price of $1011 as of Oct 20, 2013.
Identifying the set of assumptions that justify the current market valuation of Google is just the first step. More importantly, we are interested in whether the assumptions make sense and how different is valuation after replacing some of the assumptions with more realistic numbers. Below are a few variations that we have looked at.
1) The most questionable assumption is perhaps the constant EBITDA margin of 31.84%. During the past three years, Google's EBITDA margin has been declining. During the most recent quarter of 2013Q3, the margin is below 30%. Therefore it is hardly the case that the margin decline has stopped.
If we assume that the EBITDA margin gradually declines to 25% by 2022, and keep everything else unchanged from the baseline case, the intrinsic stock price is $759 per share, 25% lower than the baseline estimate. Is the 25% EBITDA margin a reasonable long-term scenario? You be the judge.
2) If we assume that Google can keep its mojo longer and converges to the steady state in 2027 (5 years later than the baseline case), and keep the constant EBITDA margin of 31.84%, the intrinsic stock price rises to $1190. However, if the EBITDA margin is assumed to drop to 25% by 2027, the intrinsic stock price drops to $903 under the extended steady-state assumption.
3) Google currently enjoys a low corporate tax rate thanks to its international tax maneuvering. However, there is always a probability that tax policies around the world may change in an unfavorable way. If we assume that the effective tax rate increases from 20% to 35% by 2022, while keeping everything else the same as the baseline case, the intrinsic stock price drops to $842.
4) An increase of the steady-state growth rate to 4% (everything else the same as the baseline case) results in an intrinsic stock price of $1063. A decrease of the steady-state growth rate to 3% results in an intrinsic price of $964.
Overall, we infer that the $1000 price tag for Google is based on optimistic market expectation about its ability to maintain current profit margin. Based on the free cash flow analysis, last Thursday's (October 17) closing price of $888.79, before the Q3 earnings release, is perhaps a better assessment of Google's fundamental value. However, in the current market environment, optimism is prevalent in the valuation of many stocks, and Google is no more over-valued than its peers (such as Facebook (FB)).