Amgen (NASDAQ:AMGN) shares (up over 22% year-to-date) have had an impressive run this year, outpacing the S&P 500 by about 6%. Nevertheless, despite outpacing the broader market, Amgen has only matched the Health Care Sector Select Sector SPDR (NYSEARCA:XLV) (up about 22% YTD) and has greatly lagged its large-cap biotech peers. For instance, this year so far shares of Gilead Sciences (NASDAQ:GILD) (up 50% YTD), Celgene (NASDAQ:CELG) (up 57% YTD), Biogen Idec (NASDAQ:BIIB) (up 55% YTD), and Regeneron Pharmaceuticals (NASDAQ:REGN) (up 55% YTD) have all increased between 50 and 57% (see graph below). But is this divergence in performance between AMGN and other large-cap biotech stocks justified?
I understand that these other, large-cap biotech stocks deserve higher multiples (i.e. higher P/E ratios and P/S ratios) because of their higher near-term growth prospects. But I believe Amgen deserves at least some degree of additional multiple expansion in the wake of the tremendous expansions in multiples experienced by these other large-cap biotech stocks. It seems odd to me that the market continues to shun Amgen while embracing the rest of biotech. At the end of the day, Amgen is a still a biotech company and is deeply involved in developing the drugs of the future, just like Gilead, Celgene, Biogen, and Regeneron. It just so happens that Amgen is in a low-growth phase of the product development cycle. Gilead went through a similar low-growth phase between 2009 and 2011. But, for the first time ever, Amgen (with a market capitalization of $80 billion) is now worth less than Gilead (with a market capitalization of $85 billion). This seems odd to me considering that Amgen has over $17 billion in sales while Gilead has only about $10 billion in sales. Furthermore, much of Gilead's growth has come from expensive acquisitions (e.g., Gilead's $11 billion acquisition of Pharmasset for its Hepatitis C drug) while Amgen's growth has been much more organic. With over $20 billion in cash and cash equivalents, Amgen could do similar acquisitions if it needed to acquire a promising drug to improve its pipeline. None of the other large-cap biotechs have this sort of bankroll.
Putting my gripes about Amgen's share price relative to other biotechs aside, I wish to use the remainder of this article to show that the current fundamentals alone support a higher share price for Amgen. After painstakingly crunching numbers and running them through a discounted cash flow model (DCF model), I have conservatively concluded fair value for Amgen to be in the range of $117 to $126 per share. This article will discuss the variables and assumptions that went into my DCF model as well as the resulting estimate of fair value. You can view a copy of my DCF spreadsheet here and follow along as I explain some of the components of my model.
Discounted Cash Flow Model
When all is said and done, the value of any business is simply the sum of its future free cash flows discounted back to present value. Therefore, we can estimate the Amgen's value by estimating its future free cash flows and discounting them back to present value at the appropriate discount rate. In my DCF model, I first estimate the Amgen's value as a firm (meaning the value of Amgen's future cash flows available to both debt and equity stakeholders) and then subtract long-term debt and add cash to that number to derive Amgen's equity value (and ultimately the value of Amgen shares by dividing by the number of shares outstanding).
There are two important parts to any DCF model: (1) the estimate of future free cash flows and (2) the rate used to discount these cash flows back to present value. I will discuss each of these important components in further detail below. If you are viewing my DCF spreadsheet, you will notice that I use "adjusted" income and "adjusted" margin numbers in my DCF model. As I explain in more detail below, I made an adjustment to these numbers to reflect the need to capitalize R&D expenditures since Amgen's primary mode of reinvestment (and therefore growth) is through R&D.
My Estimate of Future Free Cash Flows
To give the reader some context into Amgen's ability to generate cash, in fiscal year 2012, Amgen generated $5.882 billion in operating cash flow from $17.265 billion in sales. During that same year, Amgen spent $689 million on capital expenditures and $2.39 billion on acquisitions, resulting in free cash flow of $2.803 billion ($5.193 billion if we exclude acquisitions). The year prior, in fiscal year 2011, Amgen generated $4.552 billion in operating cash flow off $15.582 billion in sales. In that year, Amgen's free cash flow of $3.851 billion was markedly higher than in 2012 because Amgen spent much less on acquisitions ($701 million vs. $2.39 billion) and slightly less on capital expenditures ($567 million vs. $689 million).
In my DCF model, the data I used was sourced from AdvFn.com, Morningstar.com, and Yahoo! Finance. If you're looking at my spreadsheet here, you can see that, for each year in the forecast period, my DCF model starts with Revenues and from there derives operating income based on my estimate of operating margins. According to Yahoo Finance, analysts estimate that Amgen's revenues will grow 4% this year and 2.5% next year. I used these analyst estimates to calculate my revenue numbers for year 1 and 2 of my model. From there, I estimated that revenues will grow at 4.49% in years 3 through 5 (2015 through 2017) based on historical growth rates and historical returns on capital. From year 6 on, I estimated that revenue growth will decline until it reaches the rate of long-term stable growth in the economy, or 1.87%, in year 10 (year 2022). I estimated adjusted operating margins to be 41.87% (36.34% unadjusted) for years 1 through 5 based on historical margins, and from there trended that number to 35.77% (30.24% unadjusted), which is lower than Amgen's lowest operating margin in the last 10 years.
I estimated Amgen's effective tax rate for years 1 through 5 to be 19.20%, which is Amgen's 5-year median effective tax rate. Starting in year 6, I trended this tax rate to an effective tax rate of 25.90% in year 10, which is Amgen's 20-year median effective tax rate. Using these tax rate estimates, I arrived at my estimate for after-tax operating income each year.
As stated above, I made some adjustments to operating income (EBIT) and after-tax operating income (EBIT(1-T)) to better account for R&D expenditures, which, according to NYU finance professor Aswath Damodaran, should be treated as capital expenditures rather than operating expenses. I won't go into detail here on how I made the adjustments-you can read a paper on its importance and how to do this here (specifically, starting on page 9). Professor Damodaran and other academics and professionals in finance says it is important to make this type of adjustment for firms that primarily reinvest and generate growth through R&D. In short, the capitalization of R&D expenses (and the resulting adjustments to operating income and margins) is necessary to get an accurate picture of a firm's profitability (i.e., ROE and ROC) and reinvestment needs (i.e., reinvestment rates).
In my model, I capitalized Amgen's R&D expenses to create an R&D asset and then amortized this asset over a 10 year period (10 years because I am assuming it takes Amgen's research and development 10 years, on average, to develop into commercial products). The capitalization and corresponding amortization of R&D expenses resulted in an upward adjustment to Amgen's operating income. However, this upward adjustment to operating income was partially offset by an upward adjustment to Amgen's reinvestment needs. The net effect of capitalizing Amgen's R&D expenses was a slightly higher estimated free cash flow each year.
I used my adjusted estimate of ROC and sales-to-capital ratio (sales/invested capital), as well as historical net reinvestment rates, to estimate the amount of net reinvestment required for Amgen to achieve the estimated growth in revenues and operating income. Historical net reinvestment rates were computed as: ((Capital Expenditures + R&D Expenses + Changes to Non-Cash Working Captial + Acquisition costs) - (Depreciation Expense + R&D Amortization Expense)). Taking net reinvestment needs into account, I estimated peak Free Cash Flow to the Firm (FCFF) of approximately $5.9 billion in year 5 (2017), tapering off to about $5 billion in year 10 (2022) when the firm is assumed to enter stable growth.
Discount Rate Used (Cost of Capital)
I computed Amgen's weighted initial average cost of capital to be 7.30% based on a cost of equity of 9.13% and a tax-adjusted cost of debt of 1.83% (2.95% unadjusted). In this article, I won't explain in much detail how I derived Amgen's cost of debt. I will merely state that my estimate is based on Amgen's interest coverage ratio of 5.58 (EBIT/Interest Expense) and default spreads typically associated with such a coverage ratio. I will instead focus how I estimated Amgen's cost of equity. You can see how I computed the cost of equity, debt, and capital in more detail by clicking the "Cost of Capital Worksheet" tab of my DCF spreadsheet.
I developed Amgen's initial cost of equity (9.13%) by use of the CAPM (Capital Asset Pricing Model) where Cost of Equity = Risk-free Rate + Beta*Equity Risk Premium. The information I used to derive the Cost of Equity is from Yahoo! Finance as well as Professor Damodaran's website. For the risk-free rate, I used the 10-year Treasury Bond rate of 1.87%.
To develop the overall equity risk premium (ERP) for Amgen, I had to weight ERPs for several different regions because Amgen operates and sells products in different regions of the world. The ERP is the premium investors demand over and above the risk-free rate to compensate them for putting their money at risk in equity investments. I obtained ERP estimates for various regions of the world and the U.S. markets from public sources. Since most of Amgen's sales come from North America ($13.441 billion, or 78%), I weighted North America's estimated implied ERof 5.7% most heavily in computing my overall ERP. Amgen's remaining sales are reported as being from "rest of the world." Although not specifically reported in Amgen's 10-K, I assumed these "rest of world" sales came mostly from Western Europe ($3 billion, or 17%), with the remainder coming from Japan ($500 million, or 3%) and rest of Asia ($324 million, or 2%). Thus, after weighting each region's ERP according to sales, I ultimately derived an overall ERP of 5.84%.
To develop a beta for Amgen, I also used data from Professor Damodaran's website. Each year, Professor Damodaran provides an average unlevered beta for each industry on his website (click on "Updated Data" link). As Amgen operates in the drug/pharmaceutical industry, I used Damodaran's drug industry unlevered beta estimate of 1.03. Using this unlevered beta, I derived a levered beta of 1.24 based on Amgen's debt levels and marginal tax rate.
Plugging my risk-free rate of 1.87%, ERP of 5.84%, and levered beta of 1.24 into the CAPM formula resulted in a cost of equity of 9.13% (1.87% + (1.24*5.84%)). Since I estimated Amgen's capital structure to be about 75% equity and 25% debt, I weighted Amgen's cost of equity and after-tax cost of debt accordingly to arrive at an overall cost of capital of 7.30% ((9.13%*.75)+(1.83%*.25)). I used this cost of capital to discount free cash flow for all years in my DCF model. Typically I will use the company-specific cost of capital to discount years 1 through 5 and then trend this rate up or down to the industry cost of capital starting in year 6. In this case, however, Amgen's cost of capital is equivalent to the industry cost of capital so I used the same cost of capital throughout my DCF model.
Based on my DCF model, I computed a fair-value range for Amgen's equity of $89.9 billion to $96.6 billion, or $117 to $126 per share, which would be roughly 15 to 16 times Amgen's 2012 operating cash flow. As of the close of trading on Friday (May 17, 2013), Amgen shares were trading just over $105. Given the fact that healthcare needs are expected to increase rapidly as baby boomers enter old age and the biotech revolution is just getting started, I don't think my fair-value estimate overly optimistic-especially when you consider that I have assumed long-term growth of only 1.87% after year 10 (after 2022). Amgen appears to be in a prime position to reap the benefits of the new biotechnology era and implementation of biologics medications, not to mention the large aging population. At current prices, I think Amgen shares are still have room to run, and will do so once the market starts to realize Amgen's true potential. Several upcoming catalysts come to mind, including the UBS Global Healthcare Conference on May 21st as well as Amgen's annual shareholder meeting on May 22nd. Amgen recently announced that it "will present data from several studies of both pipeline and marketed products at the 2013 American Society of Clinical Oncology (OTC:ASCO) Annual Meeting from May 31 to June 4 in Chicago."
Additional disclosure: I do not warrant the accuracy of any data provided in this article or in the referenced DCF spreadsheet. My fair value conclusion in this article is solely my opinion. You should not treat any opinion expressed in this article as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of the author's opinion.