Cisco (CSCO) stock price took a beat of 11% after its 2014Q1 earnings release on November 13. Many investors have since been wondering if that was too much a bloodbath, or the price has more downside to go. We look at the price movement during a slightly extended window (Nov 1 to Nov 14), and compare with intrinsic price estimates during the same window. We have the following two observations to share.
First, there was an overly enthusiastic price run-up during the two weeks prior to the earnings release, which was not justified by the change in market expectations on Cisco's fundamentals during the same period. Almost half of the 11% price drop on Nov 14 is the result of a correction to this pre-announcement run-up. Second, the price drop during the slightly extended window of Nov 1 to Nov 14 is much modest, at -5.32%. This drop can be well justified by changes in market expectation about corporate fundamentals.
As a side observation from this exercise, the free cash flow valuation approach appears to be a reasonable tool to track market price movements of such a mature technology company.
Before going into the details, first take a look at the following statistics about Cisco's stock prices and analyst consensus forecasts on the three key dates: Nov 1, Nov 13 (before earnings release), and Nov 14 (after earnings release).
Analyst consensus forecasts (source: Bloomberg)
Stock price ($)
EPS 2014Q1 ($)
EPS 2014 ($)
EPS 2015 ($)
EPS 2016 ($)
Revenue 2014Q1 ($b)
Revenue 2014 ($b)
Revenue 2015 ($b)
Revenue 2016 ($b)
Price moves vs. fundamental changes
Cisco's price move during the period of Nov 1 to Nov 14 is -5.32%. However, there was a run-up of 6.34% from Nov 1 to Nov 13, before the 10.96% plunge on Nov 14. Therefore, viewed through a slightly extended window, the price reaction was more modest.
However, the 6.34% price run-up before earnings release represents an interesting case of investor optimism. From Nov 1 to Nov 13, analysts revised downward their expectations on EPS and revenue for F1Q2014, FY2014, and FY2015. Stock price moved upward despite such downward revisions.
Cisco has a history of managing investor expectations and beating analyst forecasts. Therefore the pre-announcement price run-up might be driven by the anticipation of Cisco beating the consensus again. Indeed, we see something similar prior to the August 14 release of the FQ4 2013 earnings. There was a 3% price run-up from end of July to August 14, followed by a 7.20% drop on August 15. Both times, Cisco managed to beat the consensus EPS forecasts. The price plunges after earnings announcements were caused by downward management guidance for subsequent quarters. It is safe to say that going forward, the expectation management game is not going to guarantee a price pop at earnings announcements.
In fact, Cisco's pre-announcement price run-ups fit into a broad pattern of market behavior known as the "earnings announcement premium." Statistically, stock returns during the earnings announcement months (and especially during days prior to earnings announcements) are abnormally high. This pattern has been noted in a long list of academic studies in finance and accounting. Two recent studies on this topic can be found here and here. A prevailing explanation of this earnings announcement premium phenomenon is that investors tend to increase their attention and demand for stocks with forthcoming news, regardless of whether such news are good or bad.
Changes in intrinsic value: the free cash flow perspective
If we agree that the price run-up prior to Cisco's earnings release was driven by optimism (or increased investor attention) and part of the 11% drop on Nov 14 is a correction to this run-up, then a more relevant question to ask is whether the price reaction from Nov 1 to Nov 14 (instead of the one-day drop of 11%) can be justified by changes in fundamentals.
We use the following assumptions to perform free cash flow valuation on the company. These assumptions mechanically rely on analyst consensus forecasts (data source: Bloomberg) and data from past financial statements (data source: Google Finance). It is worth saying upfront that these assumptions are not the most informative ones (feel free to revise them and come up with your own valuation). But they turn out to track Cisco's recent market price movement well--this says something about market expectations when pricing the stock.
- On each date, we take analysts' consensus revenue forecasts for FY2014, FY2015, and FY2016. From FY2017 to FY2023, assume that Cisco's revenue growth steadily converges to a terminal rate of 3.5%, which is a mix of domestic and international GDP growth rates. Free cash flows after FY2023 are assumed to be growing perpetuity with a 3.5% growth rate.
- We assume that Cisco will be able to maintain its FY2013 EBITA margin of 27.87% in future years. From FY2010 to FY2013, CISCO's EBITA margins are 27.96%, 23.50%, 27.50% and 27.87%. They are pretty stable except for FY2011.
- We assume a constant ratio of Long-term assets/Sales, at the FY2013 number of 73.38%. Long-term assets (total assets - current assets) are forecasted based on this ratio and the revenue forecasts. Investment cash flows are then inferred from long-term assets changes and depreciations.
- The depreciation rate and the Net working capital/Sales ratio are kept at the FY2013 numbers of 6.59% and 3.03%, respectively.
- Assume an effective tax rate of 35%. This is a judgment call subject to debate in any valuation exercise. Cisco's statement rate rates fluctuated from 11% (FY2013) to 21% (FY2012) in the past four years. Yes the tax rate on profits generated overseas is low; however, shareholders won't be able to get their hands on such profits without paying up to the U.S. tax rate.
- The following assumptions are made to determine the discount rate. Stock beta is 2.33 in FY2014 and is assumed to steadily reduce to 1 by FY2023 (Cisco's stock beta on Yahoo Finance is 1.32. However, a large part of equity value is due to its $50 billion cash holdings. Adjusting for cash, the beta becomes 2.33). The riskfree rate is 4%, and equity risk premium is 6%. The ratio of debt to enterprise value is assumed to be 20%, and the cost of debt is assumed to be 6%. These assumptions translate into a discount rate of 15.15% for FY2014 and a graduate decrease to b.78% by FY2023.
A copy of the valuation spreadsheet can be found here. Below is a summary of what comes out of the valuation model.
On Nov 1, based on the consensus forecasts available on that day, the intrinsic enterprise value is $103.51 billion. Adjusting for cash and debt reported for FY2013, the intrinsic equity value is $130.95 billion. Given a diluted share base of 5.43 billion, the intrinsic stock price is $24.12.
On Nov 14, based on the downward-revised consensus forecasts available by then, the intrinsic enterprise value is reduced to $99.99 billion. After adjusting for cash and debt reported for F1Q2014, the intrinsic equity value is reduced to $124.81 billion, translating into an intrinsic price of $22.98 per share.
The intrinsic price estimates on Nov 1 and Nov 14 are both slightly higher than the market prices on the two dates. This could be either because the assumptions we make are not perfectly aligned with market expectations, or because the stock is slightly mispriced. But more importantly, as long as we keep a consistent set of assumptions, the free cash valuation model should be able to offer a reasonable assessment on the change in intrinsic values given the changes in expectations about Cisco's fundamentals.
Indeed, the change in the intrinsic stock price from Nov 1 to Nov 14 is -4.70%, not too different from the change in market price, -5.32%.