Long-term treasury management
0As a consultant, I’m regularly asked questions like “how long should we borrow for?”, “should we be repaying our debt?” and “should we be making long-term investments?” One of the questions I ask in response is “what does your long-term cash flow forecast look like?” The reply is often a variation of “well, we’re not entirely sure …”
It stands to reason that to make an informed decision about long-term borrowing and investment, you need to have some sort of long-term cash flow forecast. If you have a neutral view on interest rates, you would ideally arrange your loans to mature in years of cash surplus. But if you’ve got no idea whether payments might exceed receipts, or vice versa, in ten years’ time, and by roughly how much, you’re making treasury management decisions in the dark. I know we can’t foretell the future perfectly and it’s true that things become less certain the further forward we look, but that shouldn’t stop us making some reasonable estimates.
Near-term cash flows are usually forecast on a daily basis, and with grants, precepts, council tax, business rates, payroll and other payment runs normally on fixed dates (and often for pre-determined amounts) it’s fairly easy to build a useful day to day forecast for the next couple of months. That approach is fine for the current financial year, and can be modified slightly to prepare month by month estimates for the following year. But we need a different approach to create a long-term year on year cash flow forecast.
For a start, we can assume that the revenue account is always going to balance in accounting terms – we then just need to adjust for non-cash items. In the absence of any better information, creditors and general provisions might be assumed to rise with inflation each year; this will produce a net cash inflow to the authority as growing cash income is not reflected in growing cash expenditure until a later date. Increasing debtors, where material, will offset this.
Another major non-cash item is MRP, the minimum revenue provision for repayment of debt. Because this is charged to the revenue account but does not result in a cash payment (except for finance leases and PFI schemes), it should be treated as a net cash inflow. As MRP on past capital expenditure is calculated on a formula, including a 4% reducing balance for pre-2008 assets, it can be estimated for many years into the future with some degree of certainty.
The final revenue account adjustment is for movements to or from reserves. Cash held in specific reserves is presumably going to be spent at some point, creating a cash outflow; departments will have some plans for when this might happen. On the other hand, it may be Council policy to slowly increase general reserves year on year, which will produce a net cash inflow.
We then need to look at the capital account, where cash flows and accounting entries will be quite different from each other. Expenditure needs to be forecast when invoices are paid, some of which will be in the year following the capital plan entry; with substantial slippage there could be an even longer gap. And for capital financing, we are interested in when we receive the grants, contributions and capital receipts, not the year that we apply them against expenditure. Most authorities will have a three year capital programme; some approved schemes will run beyond that time frame and can be incorporated too. Further into the future, I would only include items that are reasonably certain, such as structural maintenance on existing assets; don’t forget that we’re not really trying to plan today’s loans around schemes that may or may not get off the ground years ahead. MRP on future expenditure will also need to be built into the revenue cash flow.
An alternative approach is to forecast future years’ balance sheets, focusing on the main line items. It can be helpful to group these into the capital financing requirement, working capital, reserves and net debt; estimate the first three and the movement in net debt will then equate to the cash flow. Non-cash items like pension deficits and revaluation reserves can be ignored in both halves of the balance sheet.
Whichever method you use, it’s never going to be 100% accurate, so don’t spend too much time trying to make it so. You’re only looking for best guesses, and figures to the nearest million will be fine for all but the smallest authorities. This exercise will help to create the liability benchmark recommended by CIPFA in their new treasury risk management toolkit. But more importantly, you’ll have a sound basis to your long-term treasury decision making, and that can’t be a bad thing.
David Green is Head of Sterling Consultancy Services, a provider of treasury management advice to local authorities and other not for profit organisations. This is the writer’s personal opinion and does not constitute investment advice. It should not be relied upon when making investment decisions.