Budgeting for service delivery: meeting the challenge

By Kevin Phillips

Delivering services effectively and efficiently is the main challenge facing any government, of any country. The task is even more challenging when, as in South Africa, the government needs to make up a decades-long service delivery backlog with limited resources.

It’s not just financial resources that are scarce: there is also a shortage of the all-important skills needed to implement and manage large programmes. Even companies in the private sector find it difficult to attract and retain skills. As a result, various government departments over the years have been plagued by the twin evils of under-spending on service delivery and over-spending on less productive activities.

Detailed, careful budgeting is one way to fix these problems. At the highest level, of course, budgets are negotiated and agreed at national level by the Treasury; but what happens at the level of individual cost centres and areas of activity? It should be quite clear, to every manager in every office, exactly how much is available to spend on salaries, travel, capital projects, training and every other line item.

Just having those budgets is still not enough, however, if those budgets aren’t being actively monitored and managed — and if managers don’t have quick, regular access to information about spending in the areas under their control. It’s no good finding out that you’ve over- or under-spent on a particular item two months after the fact.

What is needed is bottom-up budgeting: the entire organisation or every department and municipality needs to be involved in the detail of what needs to be spent, by whom and on what.

It’s not practical to do this using spreadsheets, or even specialised accounting packages: those have been designed for accounting, not budgeting. A budgeting and reporting system needs to be easy for non-accountants to use and understand, because unless line managers are empowered this way, spending can never really be brought under control.

A system that has been specifically designed to bring ordinary managers into the budgeting process should allow not just for annual budgeting, but for those budgets to be updated every month or two with accurate year-to-date figures. When your budget is properly broken down by cost centre, by account and by month, and you get accurate information back about what has actually been spent, things become very different.

For example, as the end of the financial year approaches many departments rush to spend budgets so that they don’t get a lower allocation in the following year. Instead of carefully planning and implementing projects over the entire year, we get last-minute rush jobs that often don’t really achieve anything.

Good systems can help to avoid this problem. If a budget is set per month, and actual spending is also reported on time each month, managers can see very quickly if spending on a particular area or project is falling behind. It’s even possible to freeze funds at certain points, or at least limit what can be rolled over from month to month, so that employees have the incentive to plan carefully and spend consistently.

Getting control over budgets also means knowing exactly why actual spending differs from what was budgeted — whether it’s over or under. When managers can explain their actual spending, those higher up in the budgeting chain get valuable information that can help to make future budgets more realistic. This means that budgeting systems need to give managers the ability to include explanations at the level of the individual line item.

In all of this, one thing is paramount: the system must be easy to use, and easy to understand. Many budgeting systems assume that everyone is an accountant — but in reality, this just makes them completely incomprehensible to the rest of us, and frequently the system becomes the scapegoat and the “reason” for failed budgets.

Going for the “best of breed” global standard packages doesn’t guarantee getting the right solution. What works well in New York or Paris will not necessarily work well in Polokwane or Oudtshoorn – systems need to be properly adapted to local conditions. In South Africa’s case, that means serving the needs of managers who may have little or no financial training. Our systems need to be robust, fast, easy to install and easy to use — not delicate, complex and full of fancy functionality that most people will never use. Our conditions need the bakkie of budgeting systems, not the Ferrari.