The letter to the Dutch parliament is supported by a number of underlying studies. At first glance the most striking information is perhaps a cost analysis of what the introduction and operation of accrual accounting would cost the Dutch tax payer. The table below indicates that implementation costs would likely range between 129 and 261 million Euro and additional operational costs would be anywhere between 13 and 28 million Euro per year.
Table 1: Overview of the Cost of Government-Wide Implementation of Accrual Basis Accounting
Incidental (one-off) costs | Lower Threshold | Upper Threshold |
IT systems | € 30 million | € 71 million |
Personnel | € 16 million | € 25 million |
Valuation of assets | € 11 million | € 22 million |
Change process | € 72 million | € 143 million |
Total implementation | € 129 million | € 261 million |
Structural costs | | |
Personnel | € 6 million | € 14 million |
Maintenance of IT systems | € 2 million | € 4 million |
Valuation of assets | € 5 million | € 11 million |
Total structural costs per year | € 13 million | € 28 million |
Clearly these are sizeable numbers, but still one could argue the cost argument is perhaps not that strong. On a state budget of 250 billion Euro in 2011, introduction costs would come down to say 0.1 percent of annual expenditures. Spread out over 5 years this does not seem an insurmountable amount if it were for a useful reform. Having seen cost estimates for the UK, where the introduction of accrual accounting and budgeting has taken over ten years, the Dutch estimates seem, if anything, to be on the low side.
The reviews of international experience and academic literature accompanying the letter to parliament discuss in more depth a number of important drawbacks to accrual based accounting and budgeting. First, the improved quality of information that accruals is supposed to deliver is questioned. Accruals is complex and subject to many choices, assumptions and potential mistakes (and manipulation – the Greece deficit misreporting was noticed due to the much higher cash deficit, not by the partial accruals measure of ESA95). Cash accounting, moreover, can be amended with accrual information on an as needed basis, and government balance sheets can be compiled without the introduction of accrual accounting. “Modified” cash accounting can thus provide a satisfactory view on government finances is the argument.
Secondly, parliaments and ministries of finance have been hesitant to discuss the budget in accrual terms due to the increased complexity of the accrual presentation and the parallel decrease in transparency and year to year comparability. Countries like France, the U.S., Sweden, and Romania have introduced accrual accounting to supplement cash information on budget execution, but they have quite deliberately chosen not to introduce accruals for the budget itself. Even early adopter Australia has de facto gone back to budgetary decision-making on a cash basis. Only three OECD countries, New Zealand, Denmark and the U.K. presently operate their budgets fully on an accrual basis. The use of accruals, according to some of the reviewed literature favors technocratic rather than political allocation of resources (which to some might not be a bad thing of course!).
Thirdly, accrual accounting information in countries that have made the switch seems to be hardly used by public sector managers. This has been noted, for example, by the New Zealand Auditor General. Given the prevalence of current expenditure, the difference between cash and accruals for many categories of expenditure is also quite minimal, while for large investment projects cost benefit analysis can be done outside the accounting framework. This is particularly true in federal countries like Australia, where the federal budget consist mainly of transfers (except for the Defense department).
Fourth, the knowledge on accrual accounting in the government sector is limited, while accountancy professionals can be expensive to hire on a permanent basis. This leads to accrual accounting being outsourced, implemented by consultants and thus not used very much to support regular government operations.
Finally, the international comparability that accrual accounting has promised through international standards as IPSAS and IFRS has proven to be a mirage. All countries that have adopted accrual accounting, even those claiming to use international standards, have chosen national interpretations of these standards; thus international comparability remains problematic.
Many of these criticisms of accrual accounting and budgeting are not new. The studies performed are a bit short on recommendations: what are the desirable features of a good system of modified cash budgeting?; is moving to accrual accounting while keeping cash appropriations more attractive (keeping the best of the two systems) or less (similar costs, with less benefits) than going to full acruual accounting and budgeting? Perhaps, the most important comment from the Dutch Ministry of Finance is that more evaluation of the impact of accrual accounting and budgeting is necessary. With the exception of Australia, non of the early adopters have seriously reviewed the impact of their reforms. This would seem necessary before, as the Ministry puts it, the implicit assumption is made that private sector accounting practices are superior to those used in the government sector and easily transferable.
Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.
