Cost of Capital
What is it and why should we care?
This is a special Labor Day edition of Banking & Beyond, written for finance students and younger practitioners. It covers an important topic, cost of capital, which in my experience is often misunderstood, both in the classroom and in the real world.
Cost of capital is very much in the news these days. The WSJ has recently published several articles regarding the cost of capital, addressing dynamics in both the bond market and the equity market. Read together, these articles raise interesting questions regarding the seemingly inconsistent market moves in the price of debt (down) and the price of equity (up), with significance not only for investors but also for companies and practitioners of corporate finance.
Over the past twenty months, the yield on 10-year US Treasury bonds has about tripled, adding almost 300bp to the ‘risk-free’ rate, a key determinant of the cost of capital. Credit spreads have widened on non-investment grade corporate bonds, increasing further the cost of debt for some issuers. At the same time, however, US equity markets are approaching all-time highs and the earnings yield on the S&P 500 is now down to 4%, suggesting to the WSJ at least that the so-called ‘equity risk premium’ may at or near all-time lows.
Cost of capital is a topic on which we spend a lot of time in the classroom. It is also a topic with great practical application, something of which I remind my students as I see their eyes glaze over while I lecture on the intricacies of WACC and CAPM. In the classroom, precision when calculating cost of capital matters greatly, in part so that professors can determine if their students really understand what they are supposed to have learned. But over several decades as a corporate finance practitioner, I have come to believe that cost of capital is one of those metrics for which “approximately correct” generally beats “precisely wrong”, at least in the real world.
As we will discuss further below, cost of capital is not an observable data point, like the current yield on UST bonds, credit spreads or the equity yield on the S&P 500. Because cost of capital is a forward looking measure of investors’ future return expectations, not a backward looking calculation of historical realized returns or a mathematical calculation of yield from market price, there is a limit to how precise or confident we can be in any of our estimations. And to the extent we are using cost of capital for the purpose of valuation, we must keep in mind that the uncertainty associated with projected cash flows generally impacts valuation by more than the rate at which we discount those cash flows (ie cost of capital).
But whether or not we can estimate it precisely, cost of capital plays a critical role in real world capital allocation decisions. And so it is important that we all understand what we mean by cost of capital, how it is used in practice and why it is important. This is particularly true for non-financial executives, I would suggest, if only so they don’t get bamboozled by fast-talking investment bankers and others of their ilk, who are so often mistaken but so rarely in doubt.
Let’s explore.
What is ‘cost of capital’? Cost of capital is a measure of investor return expectations, adjusted for risk, with which companies set hurdle rates used in the capital allocation process. It is a measure of opportunity cost to investors, and therefore to the companies they finance. If investors believe they can earn in the capital markets a return of say 10% on investments with a certain risk profile, then the companies they finance should only allocate investors’ capital to projects which they expect will generate a comparable or higher risk-adjusted return. Broadly speaking, the expected risk-adjusted return to the investor equals the cost of capital to the company. The word ‘capital’ in this context includes both debt and equity, and so a company’s cost of capital equals the blended cost of both debt and equity, weighted by the amount of debt and equity in the company’s capital structure. We refer to this measure of cost of capital as the Weighted Average Cost of Capital (WACC).
How is cost of capital used in practice? Companies use cost of capital estimates to set hurdle rates used in the capital allocation process. ‘Capital allocation’ in this context refers to a wide variety of capital investment decisions made by companies: building a new manufacturing plant, launching a new line of business or expanding an existing one, and buying or selling a business. Cost of capital is also relevant to a wide variety of other corporate governance decisions, for example the decision to fund a business (or a project) with debt or equity, to reinvest cash into the business or return it to shareholders in the form of dividends or share buybacks, and how to structure executive incentive compensation plans in a way which aligns the interests of management with those of shareholders.
Why is cost of capital important? Cost of capital is important because of its central role in the process of allocating capital and thereby creating and destroying economic value for companies and their owners (shareholders). Companies create economic value when they invest capital to generate financial returns in excess of the cost of capital, and they destroy value when they generate investment returns below the cost of capital. When companies generate returns on investment equal to their cost of capital, they grow in size but not in value. Companies which do not have attractive investment opportunities, for which the expected return on capital equals or exceeds the cost of capital, should return excess capital to shareholders rather than allocate it to non-value enhancing opportunities.
All great companies attempt to do this, but they don’t always succeed, as realized investment returns often fall short of forecasted returns. And many not-so-great companies regularly destroy large amounts of shareholder value in the undisciplined pursuit of growth at any cost, investing in projects (including acquisitions) which are unlikely to generate returns above the true cost of capital. As one would expect, share prices over time reflect companies’ track records of capital allocation and economic profit creation, with activist investors often lobbying underperforming companies to return excess capital to their shareholders, where it can be reinvested elsewhere in the economy rather than being burned by the target company management.
John Deere and SVA. A classic example of a great company’s use of cost of capital is John Deere & Co. Deere employs as one of its principal financial management and governance tools a metric they call Shareholder Value Added (SVA), in which cost of capital plays a central role. Deere estimates that its long-term pre-tax cost of capital is roughly 12%, which it uses as a hurdle rate in the capital allocation process and to measure economic profit for a variety of other management and governance purposes, including executive (and non-executive) compensation. When Deere generates operating profit in excess of 12%, it believes that it has created value for its shareholders, which it aims to do each year but more importantly over longer periods of time. To learn more about Deere’s use of SVA, read the description on page 15 of the Deere 2022 Annual Report.
McKinsey on Value. What Deere calls SVA is a variant of basic economic value analysis, the principals of which underlie much of what we do in corporate finance, including DCF analysis. Readers who are not familiar with the broad outlines of EVA may wish to read this short book written by Tim Koller and several of his McKinsey partners, Value: The Four Cornerstones of Corporate Finance, which provides an intuitive (non-math based) overview of the subject for those charged with managing and governing non-financial corporations. More advanced readers working as corporate finance practitioners will want to read or re-read Koller’s more extensive and better known book, Valuation: Measuring and Managing the Value of Companies, the ‘bible’ of valuation for experts in the field.
Does every company have the same cost of capital? No of course not, in large part because every company is somewhat unique in terms of risk. A cyclical manufacturing company or an airline will have a very different risk profile, and therefore a different cost of capital, than will a more stable consumer products or utility company. It is also the case that different lines of business (and even projects) within a diversified company may also have quite different costs of capital, again because of the different risk profiles. And so companies must evaluate the risks associated with particular lines of business and capital projects and take these risks into account when estimating the business or project-specific cost of capital and allocating capital across their various activities. For these companies, cost of capital is not a one size fits all proposition.
In my Corporate Financial Strategy class, we use a case study in which the old HJ Heinz Corporation wanted to estimate the cost of capital for its North American Consumer Products division. The case takes place in 2010, in the wake of the global financial crisis, at a time when interest rates were at an all-time low and the stock market had regained a significant portion of its 2008 losses. Each year, most of my students come to class prepared to discuss the HJH consolidated cost of capital, but few of them have considered the possibility that the WACC for the HJH parent company may differ somewhat from that of its NACP division. At the time of the case, HJH operated several distinct lines of business in dozens of countries across the globe, each with somewhat different business and financial dynamics. And even though HJH reported its consolidated financial results in USD, a large portion of its consolidated cash flow was actually generated outside the USA in non-USD currencies, and each of these international operations would have had its own cost of capital.
What do we mean by ‘risk’? As used in the cost of capital context, the term ‘risk’ has a rather precise and somewhat whacky (WACCy?) financial definition, which incorporates ‘market risk’ but does not include company or project-specific risks which can be mitigated by investors holding a diversified investment portfolio. And so John Deere’s 12% cost of capital was determined with reference to the broad market risks associated with Deere’s business, including for example the financial impact of economic cyclicality, but it does not reflect many other company-specific business risks, such as the risk that Deere products will fall out of favor with customers.
Of course these sorts of company or project-specific business risks are very important to the capital allocation process, often much more important than pure market risks, and must be taken into account in allocating capital. The proper way to capture the impact of these non-market business risks is through the process of forecasting future cash flows, not through tweaking the estimated cost of capital. And we generally do this by modeling various possible future cash flow scenarios, not simply a single base case, and by conducting sensitivity analysis around the key drivers of operating cash flow in each scenario. In practice, however, we often see corporate finance practitioners take a short-cut by bumping up their cost of capital estimates to capture cash flow forecasting uncertainty in lieu of conducting more thoughtful scenario analysis. This approach has little empirical or theoretical support and is best avoided where possible.
Cost of capital vs cost of funding. One or the most common and consequential mistakes I have seen in my many years of corporate finance practice arises when companies (or their advisors) confuse or conflate the concepts of ‘accounting profit’ and ‘economic profit’. Accounting profits are generated when companies invest capital and generate a return above the cost of funding. Economic profits on the other hand are generated when companies invest capital and earn a return above the cost of capital. Accounting profits increase net income, but economic profits increase shareholder value.
As an example, consider a company with a 10% cost of capital (the blended after-tax cost of debt and equity in its capital structure), which elects to fund a capital investment project with borrowed money (debt) at an after-tax cost of 3%. If the project generates an after-tax annual return on investment of 5%, it will increase net income by 2% of the amount invested, the difference between the project ROI and the after-tax interest expense of the debt used to fund the project. Net income will go up and the project will be considered ‘accretive’ to earnings. But although the project is profitable in an accounting sense, it will generate an economic loss equal to the 5% difference between cost of capital and the project ROI. And this economic loss, if it continues, will translate over time into a reduction in shareholder value, even as accounting profit continues to grow.
Estimating cost of capital. For purposes of today’s post, I will resist the professorial urge to delve deeply into the mechanics of how we estimate cost of capital, the broad outlines of which are reflected in the chart at the top of this post. For those of you who would like to get more deeply into the weeds on this subject, I recommend this excellent Morgan Stanley Cost of Capital Primer.
Equity Risk Premiums. As noted, cost of capital is by definition a forward looking estimation of investor’s future return expectations, adjusted for risk. It is not a backward looking calculation of past realized returns or return premiums. This observation is a particularly relevant to the cost of equity, where many practitioners use historical equity returns and return premiums to estimate the current cost of equity capital. Historical stock market performance may influence current investor expectations of future returns, of course, but is far from clear what the correlation is between the level of past realized returns (particularly recent returns) and expected future returns. And this is particularly true in periods when underlying financial market conditions have changed significantly, for example during and after the global financial crisis.
In estimating the cost of equity, we want to know what return investors expect to earn in the future, not what they earned in the past or even what they expected to earn in the past. But this is very hard to determine in practice, at least with a high degree of confidence or precision. As noted, past realized returns may or may not be a good guide to future returns and survey data is notoriously unreliable. So what we often do instead is attempt to derive from current capital market conditions, most notably equity market valuations, data that will help us back into a better estimate of investors’ expected return on equity and specifically the ‘equity risk (or return) premium’ (ERP).
Perhaps the best known provider of this sort of ‘implied ERP’ analysis is NYU Stern Professor Aswath Damodaran (a W&M parent), who provides a wealth of current and historical data on his website. Per Damodaran’s data, the implied US Equity Risk Premium has been fairly constant over the past ten years, ranging from 4.9% to 5.6%, even though the cost of equity and cost of capital has varied considerably during this period. Over the past ten years, the yield on 10-year UST bonds (generally considered the proxy for the US ‘risk-free rate’) has ranged from below 1% to over 4% today, a much lower range than in prior years. And so by Damodaran’s estimation the implied cost of capital in the US over the past ten years has varied between 5.5% and 10%, with the changes driven mostly by changes in the risk-free rate. His latest estimate for the US market cost of capital is 9%, consisting roughly of a 4% risk-free rate and a 5% equity risk premium.
The cost of equity and the stock market. Does a rising equity market mean the cost of equity has changed? And if so, has the cost of equity likely gone up or down? If stock prices were to move up say 20% across the board, with no underlying change in the risk-free rate, does this mean that investors expect to earn a higher or lower equity return in the future than in the past, from an investment made today? I’m not sure, but the answer to this question may be somewhat more complicated than we would perhaps like to believe.
Of course when a stock index like the S&P 500 goes up 20%, it is not generally the case that all or even most stocks in the index will have also increased by 20%. Some stocks will have gone up and some will have gone down, by varying amounts, and it may be that a large share of the change in the index will be attributable to big moves in just a handful of stocks, particularly those with the largest market values in a market-cap weighted index like the S&P. In the first quarter of 2023, for example, 90% of the gain in the S&P was reportedly generated by just ten stocks, which collectively accounted for 30% of the market value of the index. The vast majority of stocks in the index were flat or down during that same period.
Changes in the market price of equities, whether of stock indexes or individual stocks, often have very little to do with cost of capital (or cost of equity) considerations. Nvidia shares have more than tripled in price this year because of changes in market expectations (speculation) about the company’s future revenues, profits and cash flows, not because of changes in its cost of capital. As noted, a large share of this year’s gain in the S&P 500 has been due to the move in Nvidia and several other tech stocks also perceived to be AI beneficiaries. And something very similar happened during the dotcom boom of the late 1990s, when the big increase in US equity markets was driven by investor speculation regarding the impact of the internet, particularly in the tech and telecom sectors, and not by changes in the cost of equity per se.
Of course there are times when the overall equity market experiences a big move up or down in tandem with the bond market, as in 2022. At these times, it is certainly reasonable to ask whether the overall market cost of capital may have changed, and if so whether companies should adjust their internal hurdle rates accordingly. As a general matter, the answer for most companies is ‘no’, unless and until they determine that these changes in capital market conditions are likely to other than temporary. As things stand in September 2023, I think we can safely say that the risk-free rate is now much higher than it was twenty months ago, which should be reflected in our current estimates of the cost of debt and equity, even if we determine that the ERP itself has not moved much.
Cost of equity vs ROE. The ‘cost of equity’ is a forward looking estimate of investors’ future equity return expectations, which is quite distinct from the corporate level ‘return on equity’ (ROE) reported on any individual company’s balance sheet. For most non-financial corporations, the book value of equity capital recorded on the balance sheet will differ markedly from the market value of the company’s shares, particularly for companies which are expected in the future to generate very high (or low) returns on capital and for companies which have over the years returned large amounts of capital to their shareholders (reducing the book value of equity). For equity investors, future returns are measured relative to the current market price of the shares, not to the book value of the shares reported on the balance sheet. In the HJ Heinz case discussed above, HJH shares traded at something like 8-9x book value, due in large part to past dividend payments and share buybacks. As a result, the company reported an ROE of 50% or so, much higher than the 3% earnings yield on its shares or the company’s 5-8% estimated cost of equity.
The impact of financial leverage. Much is made of the impact of financial leverage (capital structure) on the cost of capital, both in the classroom and in practice. The financial economists Franco Modigliani and Merton Miller won a Nobel Prize for their work demonstrating theoretically how capital structure does and does not impact cost of capital and the economic value of companies. As explained by Modigliani & Miller, capital structure does not generally impact cost of capital (the blended cost of debt and equity) other than through government subsidies arising from the tax-deductibility of interest expense, which is why (US) companies often prefer to fund projects with debt rather than equity. Debt is generally less expensive than equity due to the divergent risk allocation between debt and equity, but when companies increase financial leverage by taking on more low-cost debt, they also increase the risk and therefore the cost of the equity in their capital structure. And according to M&M, this increase in the cost of equity should entirely offset the marginal cost impact of more debt in the capital structure, before considering the tax shield.
In the real world, however, financial leverage does sometimes seem to impact the value of companies, both up and down. As a result of Modigliani & Miller’s work, we now understand that this happens not (primarily) because of the impact of leverage on cost of capital, but rather through any impact which capital structure may have on corporate strategy, operating cash flows and/or the company’s access to capital. If capital structure does not impact a company’s strategy or its operating cash flows, and if the company can always refinance by raising debt and equity funding in unlimited amounts, quickly and with minimal transaction cost, then capital structure should not be expected to have much of an impact on economic value (other than through the tax shield associated with financial debt).
These are of course very unrealistic assumptions, typical of the academic “assume a can opener” approach to economics, but until Modigliani & Miller’s pioneering academic work in the late 1950s, even these common sense observations about capital structure, cost of capital and economic value were apparently not well understood, in the classroom or among corporate finance practitioners.
Cost of capital and valuation. We should always remember that the primary driver of economic value in any capital investment project (or company) is the operating cash flow expected to be generated in the future, and not the rate at which we discount these future cash flows. As a result, most corporate finance practitioners generally spend a lot more time on financial forecasting than they do on cost of capital estimation. Nevertheless, cost of capital remains highly relevant to valuation analysis, once we have properly forecasted the expected (probability adjusted) future cash flows.
As an example, consider a capital investment project which is expected to generate annual free cash flow (FCF) of $100 million, growing at a rate (g) of 2% per annum in perpetuity. This project is considered to have a relatively low level of market risk (ie a low ‘beta’ in CAPM world), and so the estimated cost of capital (r) is only 5%, not much higher than the current risk-free rate of 4%. Using the Gordon Growth Model, the economic value of this stream of future cash flows equals $3.3 billion. [GGM: Value = FCF/(r-g)]. If our 5% cost of capital estimate is 50bp too high, however, we will have undervalued the project by $700 million, and if our cost of capital is 50bp too low, we will have overvalued the project by $440 million. These are not insignificant valuation differences, but even here the big driver of value will be the forecasted cash flows and not the cost of capital.
In conclusion. Cost of capital is an important topic which in my experience is often misunderstood, both in the classroom and in the real world. It is one of a handful of theoretical concepts that underlie much of what we do in the practice of corporate finance, particularly in the realm of corporate valuation. But cost of capital is also one of those financial metrics which we like to believe we can estimate with a high degree of confidence and precision, which in practice is not always the case.
Corporate finance professionals tend to be intelligent and highly confident people, but many of them could benefit from a bit more humility about their ability accurately to estimate cost of capital, and as a result to value companies and capital investment projects. In this regard, it might be good if we more often emulated the good folks at John Deere & Co, who know that their longstanding 12% (pre-tax) cost of capital estimate may be a bit high, and will not in any event be accurate at all points in the capital markets cycle, but who also understand that setting an aggressive and consistent hurdle rate for capital allocation purposes will help keep them focused on what really matters in their business and to their shareholders. Which in the case of Deere & Co is financial conservatism, operational excellence and a quality product, demonstrated more or less consistently from the company’s founding in 1837. And so it is that even today, 186 years later, “nothing runs like a Deere”.
Even if their cost of capital estimate does seem a bit WACCy at times.
Links:
Investors Need to Worry about the Bond Market, WSJ, August 16, 2023