Life insurers in South Africa have been stumbling over each other to issue long-term debt. The reason? Ostensibly to reduce their WACC and generate greater value for shareholders.
Common sense tells us that this is a sensible thing to do. Companies all over the world have been using debt capital to reduce their WACC by sharing the cost of financing the business with Mr Tax Collector. (The interest payments made to holders of the debt are tax-deductible expenses for the companies that issue the debt. Dividends paid to ordinary shareholders must be paid out of net-of-tax income.)
Take the example where the pre-tax cost of debt (yield to maturity on current debt or equivalently, the annualised coupon required on newly issued debt for the debt to be placed at par value) and cost of equity (a little more complicated, since the cost of equity depends on whether the equity is retained earnings or equity freshly issued through a rights offer, but will generally be based on a CAPM or APT-type model of the return required by shareholders) are equal at, say, 10%. The generic formula for the WACC is:
WACC = (1-t)*cost of debt*w + cost of equity*(1-w)
where t is the corporate tax rate and w is the percentage of total capital (measured at current market value and not book value) contributed by debt.
Thus, if w is 50%, then our WACC = (1-29%)*10%*50% + 10%*50% = 8.55%, which is lower than the 10% cost of equity. Thus, magically, but incorporating debt financing into our capital structure, we have lowered our cost of capital and increase the value of our company to our shareholders.
Or have we?
Way back in the day, two gentlemen by the names of Modigliani and Miller spent a great deal of time analysing the impact of capital structure on the value of a company. One of their key conclusions was that as we increase the amount of debt in the capital structure, we increase the risk for shareholders. This operates through increasing financial leverage because fixed costs have increased (interest payments are fixed costs in that they don’t depend on sales revenue) and thus the sensitivity of profits to shareholders to small changes in revenue increases.
Most people would agree (although some would not, and most would not agree on what really constitutes risk or a fair return thereon) that increased riskiness of profits to shareholders would result in a higher cost of equity. Without getting into a theoretical debate over the extent to which all of this really applies, I do expect that few would disagree that increased risk would result in some level of higher cost of equity. The exact size of the increase is of secondary importance here (although an interesting discussion in itself).
So, when life insurers issue debt, do they consider the increase in risk to the remaining shareholders as a result of the increased financial gearing? Well, no. And for good reason.
The reason why there is no increase in financial leverage is because most life insurers have been “matching” the liability they now have on their balance sheet with risk-free debt or corporate paper that has approximately the same duration (a specifically calculated “average term”) as the debt they have issued. The result being that, regardless of what happens to revenue, the interest expense on the issued debt is closely offset by the coupon received on the corporate paper held as an asset. Brilliant, increase in risk for shareholders begone!
Now that we have neatly side-stepped the risk implications for shareholders of raising debt capital, we should turn out attention to one of the principle motivations for issuing debt: tax deductibility of payments to providers of capital. What sounds like a good idea on the surface tends to look a little less attractive once we scratch beneath that polished exterior. The interest / coupon payments made on the issued debt are indeed tax deductible, but the interest proceeds on the debt held as assets incurs a tax charge. If the interest payments on assets and liabilities are matched, then the tax savings and payments are matched too. Net result, no impact.
(What I have described above is a simple scenario. It is possible that through some careful footwork, there might be tax advantages through, for example, generating lots of investment income within policyholder funds to utilise assessed tax losses there to reap the benefits of tax deductible interest payments in the corporate fund. However, there isn’t enough time or space to write about that here, and my perception is that this has not been the core motivation for debt issuances anyway.)
We can also look at this situation from another angle. What would happen if a company issued debt, then purchased all the debt back? Would there be tax or WACC advantages? No cash raised, all company profits and losses accrue to shareholders, and not tax advantages that SARS would permit. Surely in this case nobody would think the transaction such a bright idea?
In effect, this is what insurers have done. To be fair, they have swapped their own default risk with that of another entity (or even a pool of other entities), but whether this sophisticated credit arbitrage is a good thing is mostly beside the point since I don’t believe this was even thoroughly considered at the time.
Now, after writing this note, I can’t help but feel I must be missing something for all these insurers to be making the same decision. What have I missed?