There has been a good deal of discussion in the press about the proposal by the Planning Commission to use part of India's foreign exchange reserves to increase public investment in infrastructure. The purpose of this article is to clarify the underlying macroeconomic issues.

I specify the problem as follows:

(i) There is an urgent need to increase investment in infrastructure in order to raise the growth rate. But there are binding political constraints on raising the required resources by fiscal tightening.

At the same time, resources are available in the form of foreign exchange reserves, whose level is currently excessive.

The problem is to find a way of harnessing part of the excess reserves to increase investment in infrastructure.

(ii) The economy is currently operating at a level of capacity of utilisation consistent with reasonably low inflation. Any policy change must preserve this situation of "internal balance".

(iii) Improving infrastructure provision calls for a broad spectrum of policies. I assume that whichever method is used to raise investment in infrastructure, the government will also make every effort to reform pricing and regulatory structures.

At the existing rate of private savings, and in the absence of fiscal space, investment can only be increased by running a higher trade deficit.

The latter in turn can be financed by running down reserves or slowing down the rate at which they are accumulated in future.

Running a higher trade deficit consistently with internal balance requires a combination of policies: a policy to increase national expenditure and a policy to switch some of it to imports (or tradable goods more generally).

Neither of these policies would be sufficient by itself. On the one hand, higher expenditure would achieve a widening of the trade deficit. But it would also increase the pressure of demand on home resources and threaten internal balance by raising prices.

Hence the need for an expenditure-switching policy to divert excess demand from the home market to additional imports (over and above the import content of increased expenditure).

On the other hand, an expenditure-switching policy too would not be sufficient in isolation.

By itself, it would generate a trade deficit but also cause an undesired domestic recession. Hence the need to combine it with a policy to boost expenditure.

As regards expenditure-switching policy, there are two possibilities: exchange rate appreciation and import liberalisation. I shall assume below that in India, the latter would be preferable.

Exchange rate appreciation would discourage exports of goods and services. This would be unwise at India's current stage of development: international experience indicates that rapid export growth is a necessary condition of rapid employment and output growth.

Accelerated import liberalisation would be a better method of switching demand because it is in any case independently desirable in order to achieve higher productivity.

As regards expenditure-boosting policy, two methods are possible: fiscal and monetary. The fiscal method would be public-sector-led and is essentially the method proposed by the Planning Commission.

(Note that increased investment does not have to be directly carried out by the public sector itself. It could take the form of "viability gap financing" in which the public sector provides funds, with appropriate conditionalities, to catalyse investments made by the private sector.)

The monetary method would be private-sector-led and is the route explicitly or implicitly favoured by critics of the Planning Commission proposal. I contrast below the implications of these two methods.

For purposes of exposition, I assume that it is desired to raise infrastructure-investment expenditure by $5 billion per year for several years and that the import content of investment is 40 per cent, i.e. $2 billion.

Suppose the exchange rate is $1=Rs 50. I assume that internal balance has to be preserved and that expenditure-switching is achieved via import liberalisation, not exchange rate appreciation.

(But my basic argument would be unaffected if exchange rate appreciation were the preferred switching instrument.)

Fiscal Policy

Public investment in infrastructure is increased by Rs 250 billion. This increases demand for imports by $2 billion, which is met by running down reserves.

But it also raises home demand by Rs 150 billion. This excess demand is diverted to imports by calibrated import liberalisation, leaving the domestic pressure of demand just sufficient to maintain internal balance.

Investment in infrastructure rises by Rs 250 billion and the trade balance worsens by $5 billion.

Adopting this method raises the fiscal deficit by Rs 250 billion. The Planning Commission's case is that the rise in the fiscal deficit would be benign because neither the price level nor the interest rate would be significantly affected.

This is surely correct. The price level would be constant to a first approximation, because the pressure of home demand would be unchanged. In practice, there would be some rise in the price of non-traded goods and some fall in the price of traded goods (due to import liberalisation).

Some aggregate price rise may remain (if non-traded goods' prices rise more than traded goods' prices fall) but this would only be a relative price change and part of the equilibrating process.

It should not lead to inflation if monetary policy maintains a non-inflationary stance.

The fiscal deficit would not crowd out private investment. Since the pressure of aggregate demand is kept constant, interest rates would not rise.

The trade deficit would, other things equal, tend to reduce the money supply but the RBI can offset this reduction by open market operations or by loosening its sterilisation policy.

Of course, there are sequencing issues involved and it may be feared that if the government borrows and spends, interest rates would be driven up initially (though they would fall back later when import liberalisation eliminates excess demand).

If so, the fiscal deficit could be monetised to begin with and the higher trade deficit allowed to remove the money supply increase in due course.

These sequencing issues are no more difficult than those the Reserve Bank of India [Get Quote] normally faces. Essentially, it has to gear monetary policy to maintaining internal balance.

The rise in the fiscal deficit runs the risk of sending a "bad signal" to the market. One response would be to make an off-budget increase in public investment, which can be done in a variety of ways.

This would be compatible with the letter but not the spirit of the FRBM. It would be sensible to recognise explicitly that the FRBM fiscal target was, in effect, being relaxed for good reasons. It is really rather obvious that the fiscal stance should not be judged independently of the balance of payments position.

If properly explained, a somewhat higher fiscal deficit would not reduce the country's credit rating, given the fact of excessive reserves.

Monetary Policy

A judgement on the PC's proposal clearly requires comparing it with the relevant alternative, which in this case would be a private-sector-led rise in investment (without any public-sector subsidy).

This route also involves a combination of a switching policy (assumed as above to be import liberalisation) and an expenditure policy. The private sector cannot be ordered to invest; in order to induce it to invest, real interest rates have to be reduced.

The obvious way to achieve that is to combine monetary expansion with import liberalisation. Monetary expansion would be calibrated to reduce interest rates sufficiently to stimulate private sector spending by Rs 250 billion.

This would raise imports directly by $2 billion (assuming an import propensity of 40 per cent). Again, excess demand for home resources of Rs 150 billion would be diverted towards imports by appropriate import liberalisation.

This method has the merit of avoiding the bad signal of a higher fiscal deficit. The problem with it is that the effect of an interest rate reduction on investment is uncertain in timing and magnitude.

There is no guarantee that there would not instead be a consumption boom; and if investment is stimulated, it is not certain that it would directed to the infrastructure sectors.

Even if pricing and regulatory reforms were forthcoming, it is entirely possible that private investment may be slow to respond.

The crux of the matter can be put as follows. Using foreign exchange reserves to raise investment in infrastructure can be done by combining import liberalisation with either higher public investment, or higher private investment.

The former requires a (benign) rise in the fiscal deficit but it may have an adverse signalling effect.

The latter acts via a fall in interest rates but it would be uncertain in its impact and may not quickly deliver investment in infrastructure activities, especially those that do not normally attract sufficient private sector interest, such as rural roads, irrigation and power.

The choice between these two methods requires political and economic judgements in which the prime minister and the finance minister will be closely involved.

For what it is worth, my opinion is that in present circumstances, and on balance, the fiscal/public investment route is to be preferred.