Determining cost of capital can be a tricky matter for not-for-profits
A question that tends to be top of mind for healthcare finance leaders is, “What is our cost of capital?”
This important question would seem to have a straightforward answer. But like many such questions, the answer requires context and knowledge. One must know, for instance:
- The difference between cost of capital and hurdle rate
- How cost of capital is determined
- How not-for-profits should think about taxes
- The definitions of net present value (NPV), internal rate of return (IRR) and terminal value (TV), and how one should use them (See the sidebar below.)
Finance leaders of not-for-profit hospitals and health systems should care about this topic for two important reasons:
- Return on capital is what was achieved rather than what was planned when using the appropriate cost of capital or hurdle rate, and return on capital in excess of cost of capital is what growth is made of.
- Getting cost of capital wrong, and possibly overpaying, may put an organization at risk of losing its tax-exempt status if that overpayment is found to be private inurement.
Project risk as a determining factor in choosing which rate to use
The key questions health system financial executives should ask is are:
- For what purpose will the cost-of-capital or hurdle-rate figure be used?
- Is the project considered a core component of the existing, hospital-based business enterprise?
- Or should we use a rate different from our cost of capital, such as a hurdle rate or a rate appropriate to the risk of the project?
For example, cost of capital may not be appropriate for evaluating a new, nonhospital ambulatory surgery center, imaging center or physician acquisition and development program because the risks of those projects are fundamentally and demonstrably different from and higher than the risks for the core hospital-based business. Alternatively, using cost of capital could be appropriate for a hospital-based service maintenance or enhancement project. Finally, it may be appropriate to use a higher hurdle rate if the organization’s goal is to add financial value and growth.
In short, the decision whether to apply the organization’s cost of capital, a risk-adjusted cost of capital or a policy-based hurdle rate is up to each organization based on its mission and goals.
Calculating cost of capital
The truth is that capital cost can really be anything an organization wants it to be, as long as the organization is prepared for the consequences of that choice.
For example, one could use a low cost of capital or hurdle rate, where many projects can meet the threshold. The result would be that lots would get done, and many project sponsors would be happy. But cash could decline, margins could erode and earnings growth could reverse, possibly giving rise to compounding financial distress.
By contrast, a high cost of capital or hurdle rate would mean that much less would get done, which might be good or bad, depending on the organization’s perspective
In either case, we are discussing only the cost of capital and not the integrity of the financial forecasts that accompany any project plan.
For a not-for-profit hospital or health system, the fund balance and the claim on income that grows that fund balance are analogous to the equity of a publicly traded or private, for-profit company. The only meaningful difference is that instead of shareholders, not-for-profit hospitals and health systems have communities and stakeholders to whom they owe many important things, including financial return expressed in the form of better service, better cost and better health outcomes.
An objective view
There is a more objective way to assess cost of capital. Capital cost assessments rely on the market for publicly traded equity securities and the capital asset pricing model (CAPM), which equates shareholders’ required rate of return on capital with the volatility, or risk, of that return.
CAPM, which calculates an enterprise’s cost of equity capital (Ke), is then used to calculate a business’s weighted average cost of capital (WACC), which includes the market values of both equity and net debt (e.g., debt plus preferred stock plus minority interest less cash and investments) and its associated cost or interest rate. WACC is applied to operating cash flows derived from a business or project (in all cases excluding investment and interest income and expense but for insurance services) to determine net present value (NPV).
There is not space here to go into the details of CAPM. But it should be noted that the average cost of capital for a large, well-diversified, regional not-for profit health system with more cash and investments than debt today ranges from about 7% to 10%. The amount includes an add-on to account for the fact that most publicly traded securities are highly liquid and that most not-for-profit health systems, large as they may be, are nowhere near as large as most publicly traded, for-profit systems.
Consideration should also be given to important elements such as how competitive the market is and the type of business being considered. The following illustrates how cost of capital might differ, on average, by the type of entity.
Sample cost-of-capital calculations
|Publicly traded healthcare companies||Not-for-profit health systems||Hospitals||Ambulatory surgery centers||Physician group practices|
|Unleveraged cost of equity||9.1%||9.1%||9.1%||9.1%||9.1%|
|Leverage (net debt/capital)||70%||10%||50%||10%||5%|
|Cost of equity (at actual leverage)||13.3%||6.0%||8.9%||6.0%||5.8%|
|After-tax cost of debt||3.5%||3.5%||4.5%||5.0%||7.0%|
|Weighted average cost of capital||6.4%||5.7%||6.7%||5.9%||5.8%|
|Illiquidity and size premium||0.0%||3.0%||6.0%||9.0%||12.0%|
|Adjusted cost of capital||6.4%||8.7%||12.7%||14.9%||17.8%|
|Implied no-growth cash-flow multiple||15.6||11.5||7.9||6.7||5.6|
Although cost of capital might look like science, it involves considerable subjective judgment. For example, numerous studies have attempted to quantify the value discount (or rate addition) investors require when investing in smaller, less-liquid enterprises or when taking a minority interest in such enterprises. Typical discounts range from 20% to 50%. But what, exactly, is being measured and the extent to which various factors overlap are unclear. One may cite guidelines from various studies, but when using discounts, whether for illiquidity and size (equaling higher rates) or for minority interests, all three factors overlap. Here, subjective judgment, informed by experience, is critical.
What about taxes?
Many not-for-profit hospitals and health systems do not consider taxes because of the obvious fact that they pay little or no taxes. But not including taxes, either by tax-effecting cash flows or grossing up the discount rate, could be a mistake. A not-for-profit that fails to consider taxes, even if it is not exposed to them, may overpay solely by virtue of its not-for-profit status. That is probably not something an organization would want to do, because it could lead to the loss of tax exemption if the overpayment is found to be private inurement.
IRR, NPV and TV
Unless a project or business acquisition is expected to have a limited economic life, a TV should be calculated and used in calculating IRR and NPV. There are generally two approaches to this process.
The first is to apply a multiple to the forecast’s final-period EBITDA or income. This approach often is used in evaluating noncore projects, such as ambulatory surgery centers, in which acquisition multiples are well-known.
The second is to calculate TV according to the Gordon Growth Model, which is version of the better-known dividend discount model.a The Gordon Growth Model involves multiplying the final period’s cash flow by the sum of one plus a perpetuity growth rate (generally the inflation rate), and then dividing the product by the sum of the cost of capital minus the perpetuity growth rate.
Either approach works in project assessment and valuation. However, because TVs often represent the preponderance of total value, when performing a valuation based on the Income Approach (using discounted cash flow), using the Gordon Growth Model is often preferable, because not using it may simply replicate the Market Approach, in which transaction multiples from prior deals are applied.
In any case, it is often instructive to compare the two approaches and, if large differences appear, to understand why.
A better discipline
Not-for-profit hospitals and health systems may tend to comingle project-return expectations with community benefit goals. They should consider, however, evaluating projects in a way that segregates these two considerations and, if they wish, apply the nonpayment of taxes and/or reduced-return expectations as a clearly articulated and quantifiable statement of a decision to support community benefit, either within the project itself or to other projects or programs.
Sample net-present-value (NVP) calculations, with and without inclusion of taxes in the calculation
|Project pre-tax cash flow (CF)||-$30.0||0||3.2||3.8|
|Imputed income tax (@21%)||0||-0.7||-0.8|
|Project after-tax cash flow (CF)||-$30.0||0||2.5||$3.0|
|Terminal value (@2% perpetuity growth)||$30.6|
|Total project CF||-$30.0||$0.0||$2.5||$33.6|
|Discounted CF (@12%)||-$30.0||$0.0||$2.0||$23.9|
|NPV ($ millions)||-$4.1|
Embedded Value of Imputed Income Taxes
|Implied taxes included in project CF||$0.0||$0.7||$0.8|
|Terminal value (@2% perpetuity growth)||$8.1|
|Total project CF||$0.0||$0.7||$8.9|
|NPV ($ millions)||$6.9|
The exhibit above illustrates how a not-for-profit’s non-payment in income taxes may be made into an explicit calculation of stakeholder benefit even for projects with negative NPV.
For example, consider that a project or program produces a 6.5% return, and because of a 12% hurdle rate, it also produces a negative $4.1 million NPV either because an implicit tax on income was applied and/or because it could not be justified by a reasonable project/program forecast. Even in these circumstances, the organization might still wish to proceed (absent any potential inurement issue), but it would do so with the explicit recognition that the community benefit element to this project or program is $4.1 million.
In such a case, nonpayment of taxes covers the negative NPV and then some, but it also makes the community-benefit expectation explicit. One might, in fact, budget negative NPV, with appropriate support, as representing a quantified expressions of community benefit.
a The Gordon Growth Model is named after Myron Jules Gordon, and American economist and Professor Emeritus of Finance at the Rotman School of Management, University of Toronto, who developed the model in 1956 with Eli Shapiro. To read more about the model, see “Gordon Growth Model,” CFI Education Inc., 2015 to 2020.