In the glossary attached to its Budget highlights for 2007, the Ministry of Finance says that DE generally refers to 'expenses that represent a longer-term investment and result in the formation of a capitalisable asset of the Government. Examples of spending areas are the acquisition of heavy equipment as well as capitalisable assets, e.g. buildings and roads'.

Since such investments yield long-term benefits, it would be logical that, as a general principle, their costs be spread over time. Essentially this can be done by financing the current outlays on these investments to a significant extent by borrowing, and paying the interest and principal on such borrowings out of the enhanced tax revenues that the resulting higher GDP levels in the future would facilitate.

There are, of course, exceptions to this principle. For example, a case could be made that expenditures on military hardware are necessary for reasons of national security, even though they do not in and of themselves add to future GDP. They should thus be financed largely out of taxation. However, for more narrowly 'economic' developmental expenditures, the general principle should apply.

How does Singapore's Budget fare from this standpoint? It would be useful to look at actual expenditures during the 2007 fiscal year (FY2007).

According to the revised figures for FY2007, total DE was $7.1 billion. This was a major component of what was termed 'development estimates', which totalled $13.1 billion. The remaining $6 billion was financed by 'other development fund outlays', of which $4.8 billion comprised government loans (including to statutory boards and public enterprises), and $1.2 billion comprised land-related expenditure - described in the 2007 highlights as 'expenditure spent on land reclamation or land improvements that augment the value of land, and which are charged to past reserves since FY2001'. The $7.1 billion of DE per se is thus to be financed out of current revenues, mainly taxes.

As a matter of interest, it would be useful to know the source of the $4.8 billion of funds lent by the Government. Are they from the reserves managed by the Government Investment Corporation? Were the borrowers charged 'market' or concessional interest rates? Answers to these questions would facilitate a more inclusive understanding of the Government's fiscal policies.

Returning to the main issue: Is it a good idea to finance the $7.1 billion of DE from current revenues?

Government DE is sub-classified under four headings - security and external relations, social development, economic development and government administration.

To begin with, this classification is not entirely satisfactory: For example, surely expenditures on education, health, the arts, and environment and water resources - all of which come under 'social development' - have economic, and not just social, benefits?

Be that as it may, total DE of the two large economic ministries - Transport, and Trade and Industry - in FY2007 amounted to $3.15 billion. Given the highly economic nature of these ministries' expenditures - including, for example, 'higher expenditure on rail transport projects and improvements in commuter facilities and traffic management systems' (Transport) and 'Science & Technology Plan 2010, Enterprise Development Fund 3 and Formula One Project' (Trade and Industry) - it would certainly appear that there was considerable scope to finance a significant proportion of these outlays through loans, just as the Government lends to statutory boards and public enterprises.

Since a strong case can be made that much of social development has an economic dimension, part of the total DE of the six social development ministries, amounting to $2.9 billion, could also well have been financed by loans.

All this would suggest that it would not be prudent to lower personal and corporate income taxes much further, unless absolutely dictated by competitiveness concerns. Otherwise it would be difficult for the Government to obtain adequate tax revenue from future GDP growth to service the increased debt, should it finance more of its DE through borrowing, as is being advocated here

Although the Government would, effectively, be repaying the loans to itself (unless it chooses to borrow from the market), this would be sound fiscal practice, just as statutory boards/public enterprises do in fact repay their loans.

The issue arises as to whether the GST increase of 2 percentage points last year was necessary. It was estimated last year that this increase would contribute up to $900 million of gross revenue (which has turned out to be a significant under-estimate). I would argue that the increase would not have been necessary had greater recourse been taken to borrowing to finance DE.

It has been indicated that the GST increase was designed to generate more revenue in future years, not just 2007. However, taking into account the GST offsets for the next few years, the net revenue gain for these years is unlikely to be high. It should be noted too that GST-induced increases in prices reduce the real value of individuals' fixed-nominal-value assets, such as bank deposits and CPF balances, unless offset by higher interest rates.

It would have been better all round to have relied more on borrowing-financed DE, thus reducing current tax revenue needs. That debt could have been serviced in the future when people would be better off and, as a result, would be paying higher taxes even if the tax rates were unchanged. Any GST increase could then have been postponed to a later date, if such an increase became necessary then.

Technically, economists would refer to my argument as an 'efficiency argument': Policy measures that finance investments for future benefits by imposing costs in the present - when incomes are lower and many people face borrowing constraints - rather than in the future, when incomes are higher, are inefficient.

Also, to the extent that current cost-bearers and future beneficiaries are different, considerations of intergenerational equity would be involved.

Such considerations are also highly relevant in seeking to determine the desirable levels of annual fiscal surpluses, and the resulting reserve accumulation over time. A wider public discussion of, and consensus on, this would be beneficial.

The writer is Professor of Economics and Director, Singapore Centre for Applied and Policy Economics, NUS. The views expressed here are personal.