Government uses discount rates when it decides whether to invest or sell assets; regulators use discount rates when they set prices for natural monopolies; investors use discount rates when they buy, sell and invest.
Our work at the National Audit Office shows that this type of analysis is useful, but it can be very tricky to apply. This article summarises mini case studies drawn from our published reports, and concludes with some tips to improve the application and interpretation of discounted cash flow analysis.
HM Treasury advocates discounted cash flow analysis for investment appraisal in central government. It explains that discounting is based on the principle that, people prefer to receive goods and services now rather than later. Government calls this the ‘social time preference rate’ – currently a ‘real’ rate of 3.5% (a nominal rate of 6.09%).
When valuing public sector assets, supplementary guidance notes that using the Green Book discount rate will usually be higher than the private sector value because the private sector will apply a higher discount rate. This might indicate that assets should never be sold. But the guidance suggests the relative efficiency of public and private ownership may differ and states that a higher “market-based risk” discount can be used in calculation of both asset retention and disposal scenarios.
A change in ownership may affect the future cash flow. The privatisation of Royal Mail in 2013 allowed access to private capital and commercial disciplines. Government forecast that earnings would fall by approximately a quarter compared to performance in 2012-13 under continued public ownership, but would increase significantly if sold. The Department estimated the value of retaining Royal Mail in public ownership as a cash inflow for the tax payer of £0.3bn to £1bn. The government actually sold 60 per cent of the government’s shares for £1.98bn, and the market value of these shares increased by 38% on the first day of trading.
Forecasts are fallible because it is hard to predict the future. The actual number of Eurostar passengers is significantly lower than was originally forecast in 1995. It is difficult to forecast future cash flows reliably even after identifying a range of relevant factors, such as low cost airlines, investment in new trains, and the prospect of direct rail competition.
The present value of a future cash flow reduces as the discount rate is increased. The present value of the government’s preference share in Eurostar ranged from £158m to £216m depending on the discount rate applied. Eurostar redeemed it for £172m (a 9.4% discount rate).
The same valuation can be derived by a wide range of discount rates. Three valuations of Eurostar converged on a tight range - £305-£315m, despite the discount rates varying from 9.5% to 17%. Previous valuation work had used a discount rate of 8.5%.
Government is sometimes selective in whether to apply discount rates. The historical cost of financial sector interventions, such as Lloyds and RBS is not necessarily representative of their present value. However the overall cost policy decisions involves opportunity costs which can be measured using the social time preference rate.
When economic regulators set price controls for monopolies, such as in the water sector, the capital asset pricing model (CAPM) is used routinely to estimate the cost of capital.
Private equity ownership has largely replaced the stock market, so it can be difficult to estimate beta where there is a lack of directly comparable companies with listed shares. Gearing in the water sector has increased since privatisation. Private sector debt is generally priced at a premium to public debt to reflect risk. However, regulated firms have been able to borrow at lower rates as interest rates have fallen during the 5-year price control period. By contrast, private finance initiative projects were financed using fixed rate debt and interest rate swaps. It can be expensive to redeem bonds (which are fixed rate borrowing) or buy out interest rate swaps (contracts used to fix the interest rate for bank loans). We estimate that the swap liabilities of all PFI deals are currently around £6 billion.
Reducing the risk of bias or error will require: clarity on the key assumptions to avoid embarrassment and learn from mistakes; less ‘political’ estimates, (e.g. to justify a chosen outcome); cross-checks using a variety of techniques; sensitivity analysis and “what if” scenarios.