Recently, Intel announced the delay of its 7-nanometer integrated circuit process. Affected by the news of Intel’s delay, Intel stock moderately sold off. At the same time, Intel’s competitors AMD (Advanced Micro Devices) and Taiwan Semiconductor Manufacturing (TSMC) have seen their stock prices rise sharply. Given the recent hype in technology stocks, AMD and TSMC stocks have been very hot recently. I think Intel’s sell-off has been exaggerated, while the valuations of AMD and TSMC suggest unrealistic growth. The purpose of this article is to perform a fundamental DCF analysis of Intel and compare it with the valuations of AMD and TSMC.
Background
The narrative is that Intel’s latest chips will be delayed, their rival AMD will gain the lead by selling 7nm devices. AMD does not produce its chips. Instead, AMD signed a contract to produce its circuits with Taiwan Semiconductor Manufacturing Co (TSMC), the largest IC contract manufacturer. As another twist, Intel has stated that if they need a backup plan, they will have TSMC produce some of their chips. The market is interpreting this announcement as Intel is doomed and that AMD and TSMC will grow like crazy. This is reflected in the share price of these stocks, which is shown below.
Both AMD and TSMC stock prices have gone parabolic lately. Part of this is from Intel’s news, while a lot of it is their gains are from the recent tech stock mania.
By doing a rough DCF valuation on these companies, I believe Intel is reasonably valued. As for AMD and TSMC, I believe the market is pricing in unrealistic growth. The narrative may be true, that Intel loses market share. Maybe Intel becomes the next IBM, being less relevant than its glory days. On the flip side, I believe overpaying for a stock is far riskier than scary headlines. Value investors look for times when negative news is overblown, and the stock can be bought at a discount.
Intel DCF
First, let’s estimate the value of Intel using a discounted cash flow model. Intel’s free cash flows for the last few years, plus trailing-twelve-month, are shown in the table below. I’ll use the average of these FCF figures in my model by using $15B as the input. Over the last five years, Intel has grown revenue at 7% a year. I will handicap this a bit by using a 5% growth rate for the next ten years. For the growth rate beyond year 10, I will use a value of 2%. The last input to the model is the discount rate, which I will set to 10%. These values estimate the value of Intel at $51.20 a share. Currently, Intel is trading for around $50 a share.
| 2017 | 2018 | 2019 | TTM |
| 10.33B | 14.24B | 16.93B | 21.9B |
Based on this analysis, Intel appears to be fairly valued. It may not be a screaming bargain, but it is trading at a reasonable business valuation. I believe the market’s narrative is exaggerating Intel’s troubles. It is quite possible that you could buy Intel, and the business produces a 10% average return. As we will see, I do not believe that is the case with AMD or TSMC.
TSMC DCF
Next, we will perform the same calculation on TSMC. The table below shows TSMC’s recent free cash flow figures. An average free cash flow value of $7.3B seems reasonable, so that will be input into the model. In the last five years, the income growth rate is 8%, so we assume that the cash flow in the next ten years can grow at this rate. The terminal growth rate and discount rate are the same as the Intel example. These DCF inputs yield an estimated value of $27.25, while TSMC currently trades around $79 a share.
| 2017 | 2018 | 2019 | TTM |
| 8.56B | 8.41B | 5.16B | 5.16B |
As you can see, there is a significant discrepancy between my estimate of what TSMC’s business is worth compared to what the market thinks. Perhaps I underrated TSMC’s growth rate since the narrative is that it will gain market share on Intel. Let’s try redoing the calculation by using a 15% free cash flow growth rate instead of 8%. This produces a value of $42.80 per share. Ok, how about we try an even more optimistic growth rate of 20%. The DCF outputs a value of $59.15 per share for TSMC, still significantly below what it currently trades at.
Another technique I like using when looking at stocks is performing a reverse DCF. This is done by specifying the current cash flow, discount rate, and current stock price. Then address the growth rate instead of the business value. This can help investors understand what kind of growth the market is pricing in stocks. Through this exercise, based on TSMC’s current prices, the market estimates a ten-year growth rate of 25%.
I believe a 25% growth rate for TSMC is rather optimistic. Many companies can grow at high rates for short bursts, but it is difficult to sustain greater than 20% growth for ten years. It is possible TSMC could pull it off, but I assign a pretty low probability to this.
AMD DCF
Finally, in our Intel 3-way show off, we will calculate the value of AMD. The recent cash flows are shown below, and we can see 2017 and 2018 cash flows are negative. The 2019 FCF figure is slightly positive, and the TTM number is the highest in this series of data. To be generous, we use the round number of $600M. In this example, I will use a 7% growth rate, which is why AMD’s revenue has been growing in the last five years. With these inputs, the DCF produces a whopping $9.70 for AMD’s shares…AMD currently trades at $78.
| 2017 | 2018 | 2019 | TTM |
| -101M | -129M | 276M | 611M |
To give AMD the benefit of the doubt, let’s recalculate its value using higher growth rates. Using a 15% growth rate produces a value of $15.95, while a 20% growth rate yields $21.90 a share. Both of these are well below what AMD currently trades at.
Using the reverse DCF, we can see how much growth the market is pricing in. This technique shows that AMD would have to grow at 40% per year for ten years straight to justify its value. This growth rate is completely unrealistic. Sure, the stock price of AMD has skyrocketed lately, but the business hardly makes any money. The price of AMD makes no sense when looking at it as a business. Instead, the market is valuing AMD based on its narratives, making increased profits from AI, self-driving cars, cryptocurrency mining, and, most recently, gaining market share from Intel. These could all come true, but an investor must pay a reasonable price for growth.
Conclusion
As a value investor, I try to find situations where the market hates a stock or industry. Usually, these narratives are exaggerated, which creates opportunities. The market has also exaggerated potential growth, which has resulted in stock transactions that are much higher than actual growth rates. With this post, I tried to show that the market is pessimistic towards Intel. Still, it is trading at a decent business valuation. Intel’s competitors, however, are trading at values that imply unrealistic growth. AMD and TSMC may continue to do well in the short term, but valuations matter over the long term.
For more value investing content check out:
The Many Flavors of Value Investing