A quarterly magazine of the IMF

December 2002, Volume 39, Number 4

The VAT in practice

The qualifications and negatives of the VAT, in principle, described by Ebrill, Keen, Bodin, and Summers in their otherwise excellent article, "The Allure of the Value-Added Tax" (June 2002), do not fully and fairly sum up the VAT's demerits in practice, judging from Australia's experience with our VAT equivalent, the Goods and Services Tax (GST).

The GST has horrendous compliance costs for the two to three million smaller businesses now burdened with implementing it, whereas its predecessor, the Wholesales Tax, was imposed on fewer than 100,000 businesses. The GST's huge compliance costs are rarely acknowledged by officials, academics, and big businesses that have the luxury of appraising the tax in principle, like your authors. The onerous compliance costs stem from the need of the taxpayer to examine virtually each and every item of business expenditure and revenue for its tax liability, an utterly unproductive tax self-assessment that smaller businesses cannot afford to entrust to junior or inexperienced staff.

Under the GST, tax self-assessment has become almost an end in itself, tending to edge out the primary objective of business to earn income. The stringent compliance requirements of the GST, in my opinion and that of many accountants and other professionals, threaten the very integrity of the Australian indirect tax system.

The task ahead in Australia now is to reform the tax system, which had been the very political motive for introducing the GST originally.

It would be unfortunate if the enthusiasm of your authors for the VAT induced unsuspecting officials who underestimate the difficulties of complying with the tax to introduce it.

John P. McAuley,
Turramurra, New South Wales, Australia

The issue of compliance costs is an important one, and it is discussed at some length in Chapter 5 of the book, The Modern VAT, on which the article is based.

While I cannot comment on the Australian experience, I would stress two general points. The first is that, in many developing countries, the tax that the VAT replaces has often been far from easy to comply with (or administer), commonly involving multiple rates, exemptions, and partial crediting arrangements. Though there is little firm evidence to draw on, a simple VAT is often inherently less complex than what it replaces. It may be that the self-assessment procedures associated with the VAT are unfamiliar to some taxpayers, but part of the wider purpose of the reform is precisely to strengthen the overall tax system by developing this way of doing business. Second, awareness of the potential compliance difficulties of smaller taxpayers is one of the key reasons why the IMF advocates setting a relatively high threshold for the VAT, so that those who might find the tax most difficult to comply with simply do not have to.

In "Moral Hazard in IMF Loans: How Big a Concern?" (September 2002), Kenneth Rogoff argues against his own theory by saying that the moral hazard in IMF loans is not a big concern. He bases this on the fact that the vast majority of IMF loans are repaid.

However, a more sophisticated theory of moral hazard goes like this. Debtor countries get into repayment difficulties because they expect the IMF to bail them out in the event of trouble. Because IMF approval acts as leverage to raise private capital, debtors make sure that they repay IMF loans first. So, when trouble arises, private creditors are left holding the baby while the IMF comes out unscathed.

Instead of taxpayers bearing the burden, as in the original theory, the burden is borne by private creditors. But because these creditors are also taxpayers, it comes to the same thing.

The fact that the IMF (and the World Bank) gets its loans repaid is no argument against the theory of moral hazard.

Michael Beenstock

I certainly did not assert that moral hazard has never been, and never will be, a concern. So far, and things could change rapidly enough, serious researchers have found it hard to definitively prove its importance empirically.

In addition, if it were true that IMF lending systematically benefits countries at the expense of private creditors (something most proponents of moral hazard would deny), then this would be reflected in higher borrowing costs and, as a result, could not be a source of debtor country moral hazard. So this "theory of moral hazard" does not make much sense.