We use actual rather than notional tax charges
Distinctive approach: Pier Analysis doesn't make a widely used adjustment to rates of return
This note is part of the Pier Review reference material.
Cost of capital
First claim on the cash flows generated by an enterprise is held by any institution that has lent it money. Once the lenders are satisfied, any residue flows to the investors who own the business.
Since these two constituencies share a claim on this enterprise cash flow, business schools teach students to discount future enterprise cash flows at a cost of capital which blends
the cost of the debt, which is essentially the interest rate
the cost of the equity, which is derived using the Capital Asset Pricing Model.
The two costs are weighted in proportion to the amounts of debt and equity outstanding. As a result, it is known as the WACC, or the weighted average cost of capital.
We could summarise this in a formula:
WACC = %equity x cost of equity + %debt x cost of debt
But this isn’t the formula taught in business schools. They invite students to modify the formula, by putting this bracketed term at the end:
WACC = %equity x cost of equity + %debt x cost of debt x (1 - t)
Here t is the corporate tax rate. In the UK at the moment, it is 19%. So the cost of debt, in other words the interest rate, is to be scaled by 81%.
The argument for doing this is that in nearly all jurisdictions companies are allowed to deduct their interest expenses when calculating the profits on which corporate tax is levied. To this extent, the interest cost is smaller than implied by the headline rate, because this mechanism is providing what is often called a tax shield.
Because the benefit of the tax shield is captured by reducing the cost of debt, it should not be counted again by including it in the tax calculation. MBA students are taught to calculate a notional tax charge, also known as an ungeared tax charge, which doesn’t make a deduction for interest, though it does include all the other expenses that are allowable against tax by the jurisdiction in which the firm is operating.
Take care with the shortcut
This 1 – t adjustment is designed to save the effort of working out what the company’s interest and tax bill will be in future years. It is a short cut, but it is so taught that generations of MBAs parrot it as if it is the One True Way of performing the calculation. [1]
The difficulty is that it is a short cut that is only safe to take if it truly is the case that a company can get a tax deduction for its interest costs. It works for mature companies with moderate debt in a steady state, but it is not applicable for heavily borrowed companies near the start of their development.
Infrastructure precedent
Investors in infrastructure would, or at least should, never take this short cut because it can take years before their projects become fully tax paying. Their projects are generally characterised by
high interest costs, which fall gradually over time as they repay the large borrowings they take out to fund the construction of a substantial asset
revenues which are not very large at the start, though they rise over time as production ramps up and sales prices rise with inflation.
Gross margins are small or even negative in the early years, and certainly not enough to cover all of the interest bill. As a result, it is simply not the case that the interest burden is effectively shielded by the tax charge. [2]
Similar considerations apply in real estate, and in technology businesses, for which a decade may pass before they pay any tax. [3] Famously, there are technology ventures which are among the largest companies in the world, yet pay little tax.
The consequence of making the 1 – t adjustment when there is no tax shield for interest, or it exists but only after several years’ delay, is to underestimate the cost of debt. That feeds through to the WACC, causing underdiscounting and overestimation of valuations. At least some of the current excess valuation of the stock market may be attributed to this cause.
Infrastructure investors will develop and share a financial model, which goes to the trouble of doing the work which the 1 – t adjustment is aiming to sidestep. Their model will calculate the interest and tax liabilities carefully in every year of the project. This being so, there is no reason to take the 1 - t short cut.
Models of infrastructure projects prepared by individuals with a poor grasp of the reasoning behind the 1 – t adjustment quite often go to the trouble of performing two tax calculations
the notional ungeared one, just described, from which are derived cash flows for discounting
the real, or geared version giving a more realistic indication of what the venture will actually pay; it is needed to forecast the firm’s cash flows, and the rate at which it can service its debt.
It’s ironic that the 1 – t adjustment, which was intended to be a labour saving short cut, here results in doubling the workload, and the chance of confusion and error in the model.
The more usual approach in infrastructure finance is not to make the 1 – t adjustment, but to use the geared discount rate on the enterprise cash flows, before financing but after geared tax. There’s just one tax calculation, the one that reflects the firm’s true tax position.
This is the approach that Pier Analysis uses in its modelling.
[1] The extent to which this approach has been embedded in courses, text books and online materials as the norm is extraordinary. It probably helps that the original work in this field won its authors a noble prize.
[2] To be fair, most technology businesses would not be able to raise significant borrowings, being entirely funding by risk equity, so the cost of debt, and any adjustment to it, would be irrelevant.
[3] International efforts to stop Base Erosion and Profit Shifting go some way to reducing the extent to which interest deductions can eliminate or defer tax liabilities.