ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

The Critics Are Not Keynesian Enough

The Critics Are Not Keynesian Enough

It is because fiscal policy-makers may not respect intertemporal constraints that it is better to bind ourselves to rules. Also, in an intertemporal set-up, lowering fiscal deficits can be expansionary.

The Critics Are Not Keynesian Enough

Fiscal Responsibility and Growth

It is because fiscal policy-makers may not respect intertemporal constraints that it is better to bind ourselves to rules. Also, in an intertemporal set-up, lowering fiscal deficits can be expansionary.


hose who favour discretionary fiscal policy for the purposes of stabilising output over the business cycle see those who ask for limits to fiscal deficits as being unusually fetishistic about fiscal policy. This labelling is blind to the nuance that we can only fail to recognise at our own peril – that fiscal policy which has historically been associated with the short-run role of stabilising output has increasingly come to adopt a long-run objective as well of fiscal discipline due to the ballooning of debt and deficits.1 Fiscal discipline means that the government faces an intertemporal budget constraint and any fiscal action taken by the government does not violate this constraint. This does not require the doing away of deficits altogether. Some deficits may be caused by government expenditure that is an investment expenditure which is growth promoting and so may well pay for itself through the taxes garnered from the increase in output. Such deficits are not a hazard and it is difficult to label them as problematic for long-term fiscal discipline. Similarly a few years of deficits is not antithetical to fiscal discipline provided they are followed by a few years of surpluses. However, when future action to reverse a sequence of deficits is not predictable, a government may be seen as not respecting its intertemporal budget constraint and thereby reneging on fiscal discipline. Moreover, as governments are in office for limited time spans and they cannot commit to deficit reduction on behalf of their successors, there is a built in bias for deficits as the burden of the deficit is not faced by the incumbent government. Hence it is important that whilst governments have the short-run flexibility to stabilise the economy they also respect the long-run objective of fiscal discipline as given by their intertemporal budget constraints.

The belief in the room for fiscal manoeuvre in the face of a cyclical downturn is based on the presumption that the debt is sustainable. The empirical story does not support this presumption2 . In fact, with interest payments on the debt rising, greater proportions of the revenues of the government are being diverted to interest payments and these transfers are being made at the cost of productive public expenditures. This requires fiscal correction and delay will just cause financial markets to hike interest rates in anticipation of the lack of sustainability which will affect the government’s ability to intervene in the economy. It is due to this that the concerns regarding fiscal policy have been on longterm issues around fiscal discipline rather than on short-term issues regarding aggregate demand management.

A formal analysis of the dynamics of debt makes clear that a few years of debt decline is not enough to attain sustainability. What may be needed is a procedure that guarantees year after year that the debt-to-GDP ratio is stabilised. That is why the Fiscal Responsibility and Budget Management Act (FRBMA). There are two important points to note about the intertemporal budget constraint of the government. First, it is not a behavioural equation and so it is silent on the question of which of the government’s revenue and expenditure variables should be adjusted in the face of disequilibrium. Second, the intertemporal budget constraint is also silent on the question of the timing of any required adjustment and simply specifies that an adjustment must occur, sooner or later. In practice, this keeps open the possibility of delay in undertaking what could be a costly adjustment programme for the government and deficits may just balloon. To keep out this possibility governments have been enacting rules that set strict limits on the size of deficits. Chile and Brazil, for instance, have set strict limits on the budget deficit. In Chile the rule in fact calls for a structural surplus of 1 per cent of GDP. The Brazilian rule establishes that the congress sets limits to spending on personnel over a three-year horizon while the government commits to a pre-announced primary budget balance. In India, Parliament passed the FRBMA in August 2003 which requires the fiscal deficit of the central government to not exceed 3 per cent of GDP by 2007-08 and that the deficit on the revenue account be eliminated by the same date.

Inflexible Rules

In all cases where such acts have been passed, the limits to the deficit are selfimposed and there are no sanctions when the deficit limits are breached. As the rules are generally non-binding there are chances that they will be set aside when economic and political expediency require and this has already occurred in India last year (2005-06) when the target for the revenue deficit was put into “pause” mode by the finance minister on account of a higher demand on resources arising out of the award of the Twelfth Finance Commission. A problem with rules for deficits then is that they are not flexible enough in the face of unexpected events. This debilitates the discipline that the FRBMA is designed to deliver. Rules like the FRBMA work to promote the long-run quest for discipline in spending but at the cost of ignoring short-run exigencies. The limit of 3 per cent of GDP on the fiscal deficit is also somewhat arbitrary and seems to have been selected from the same limit put for European Monetary Union membership in the European Union’s Stability and Growth Pact.3

This means that it is considered alright if public investments – presumed to average 3 per cent of GDP – are financed through borrowing. Thus, deficits are tolerated as long as they do not exceed the size of public investment spending. However, there is no guarantee that public investments generate a rate of return that matches the cost of borrowing. Also there is no economic rationale for what in budgetary accounting is referred to as public investment. Education is classified as a public consumption expenditure whilst the building of new government offices is considered an investment. However, public spending on education is likely to be far more socially productive than building a new government building. Rules

Economic and Political Weekly November 4, 2006 then do not leave room for flexibility in the face of unexpected events. Also rules do not matter much if they are not backed by an enforcement mechanism. This leaves the task of deficit containment in the face of unsustainable levels of debt to the presence of competent and dedicated policy-makers and politicians who have good judgment and know when to exercise discretion and when to press for fiscal discipline. Of course, that requires developing institutions where incentives exist to act in such a way and until we devise such institutions it would be preferable to bind ourselves to a deficit rule as has been enacted.

Critics Are Fundamentalists

However, having said that we now go on to argue that those who advance the label of fiscal fetishism to those who argue for deficit rules are actually fiscal fundamentalists who prefer to work with a shortrun framework in which reductions in fiscal deficits are necessarily contractionary. However, in an intermediate run framework where the intertemporal effects have had a chance to play themselves out just the opposite is true – a reduction in fiscal deficits can be expansionary as we demonstrate. Second, most writing on the effects of fiscal deficits implicitly work in a closed economy framework and ignores its twin

  • the current account deficit which in macroeconomics is known to be related to the fiscal deficit. This oversight is also addressed by us. In short, the critics of deficit targets are not Keynesian enough
  • i e, they do not pay attention enough to the process by which output and expenditure are determined. When we look at fiscal policy in an intertemporal open economy framework a decline in the fiscal deficit is expansionary. So whether we have deficit rules or we rely on good judgment by policy-makers is an operational issue. The more substantive issue is that a reduction in deficits at the current juncture is good economic policy.
  • That additions to capital stock augment aggregate output on the supply side is not even considered in the standard approach to income determination. The presumption usually is that the economy has underutilised capacity and so national output adjusts endogenously to changes in aggregate demand. We use a two-period intertemporal approach in the framework of aggregate expenditure and aggregate output.

    Macroeconomic equilibrium in the market for goods as we know is where aggregate output equals aggregate expenditure. Aggregate expenditure in turn is the sum of domestic expenditure or absorption, C + I + G, and the current account balance, CAB = (X – M) + NFI, where NFI is net factor incomes from abroad. We assume real interest rate parity4and perfect capital mobility. Moreover, we also assume static inflation expectations and exchange rate expectations and ignore other risk factors that restrict capital mobility. Under these conditions the exogenous world interest rate r* is equal to the domestic interest rate, or,

    r = r* …(1)

    Consumption is a function of disposable income

    C = c0 + c1 (Yn – T) …(2)

    Similarly, with static exchange rate expectations the competitiveness effect of exchange rates on exports and imports are temporarily downplayed and imports respond to aggregate domestic income whereas exports respond to the exogenous aggregate income in the rest of the world. In an open economy the relevant aggregate output and income variable is GNP which is GNP = GDP + Net Factor Incomes from

    Abroad Net Factor Incomes from abroad is simply the interest receipts on net foreign assets held, or, r*F where F is the stock of foreign assets. Thus,

    Yn = Y + r*F …(3)

    We begin the analysis at the point of time where the initial pre-existing stock of foreign assets is zero, F = 0. In that case Yn = Y, national income equals aggregate domestic income, and the current account balance equals the trade balance.5

    In the initial period of analysis, then, if the current account is in deficit, (X – M) < 0, then, the excess imports must be financed through a capital account surplus or a capital inflow – a net inflow of foreign direct investment (FDI), portfolio investment, and banking capital. The capital account balance is none other than the net foreign asset position of the nation’s residents vis-à-vis the rest of the world. Hence, after the initial period, the capital account balance (KAB) is capital inflows net of outflows and this is the net accretion to the nation’s stock of foreign assets, i e,

    – KAB = ΔF = CAB …(4)

    In an intermediate-run model the aggregate output produced is a function of capacity creation via additions to capital stock. With a given labour force N , the output produced Q is a function of the capital stock, or,

    Q =F(K ,N) =f (K1) …(5)


    Increments to capital stock result in increments to output but at a diminishing marginal rate. Ignoring depreciation of the capital stock, any additional investment adds to capital stock,

    K =K +I


    The additional capital stock resulting from investment expenditure enlarges national output and income in a subsequent period. A marginal addition to capital stock results in an increment to product or output as given by the marginal product of capital,

    f ′(). Hence, if the capital stock

    MPK = K increases by ΔK = I units, the increment to output is approximately ΔQ = MPK(I). We may then write

    Qt+1 = Qt + MPK(It) …(6)

    The output produced results in income for the agents in the economy which is either consumed or saved. For the private sector, saving is disposable income (Yn – T) less consumption, or,

    Spvt = Yn – T – C

    Saving by the government is given by tax receipts less outlays by the government, or,

    Sgovt = T – G

    Total national saving is then given by

    S = Spvt + Sgovt = (Yn – T – C) + (T – G)

    = Yn– C – G …(7)

    The aggregate expenditure equation is given by

    Yn =C +I +G +(X −M +NFI)

    or, Yn

    −C −G =I +CAB where, the current account balance is CAB = X – M + NFI. Substituting for the left-hand side of the above from (7), S = I + CAB

    or, S – CAB = I …(8)

    Thus, domestic savings and the savings of foreigners6 finance investment expenditures in the economy. This is nothing but the familiar condition that the market for output is in equilibrium when aggregate savings equals aggregate investment.

    Economic and Political Weekly November 4, 2006

    Figure 1: Macroeconomic Equilibrium in an Open Economy


    Yt Yt+1/

    nn OutputQt+1

    We now bring together all this in a single the 45° line. To the consumption function diagram – Figure 1 – where the horizontal add another component of aggregate axis measures national output or the output domestic expenditure which is governof final goods and services made available ment expenditure, G. As government for sale in the domestic market and abroad. expenditure is exogenous and taken to This is because part of the output unlike be public consumption expenditure, the in a closed economy is sold to meet the C(Yn) + G schedule is parallel to the C demand for exports. schedule as government expenditure is con

    stant. If the national output produced in

    Q = Yn + M …(9) n

    the initial time period is Y = Q , then,

    tt On the vertical axis we measure aggre-from (7) national savings is given by gate domestic expenditure E which com-St=Ytn– C(Yn)t– Gt. St or national savings


    prises demand for domestic goods and is the difference between the value of Yt demand for foreign goods or imports – as measured on the vertical axis via the 45°

    line and the C(Yn) + G schedule.

    E = (C + I + G) + M …(10)

    We now map out the investment sched-

    Now, suppose in the initial period where ule. Recall that investment multiplied by there is no pre-existing stock of financial the marginal productivity of capital results assets aggregate domestic expenditure is in the incremental output produced by the less than output. Then, with E<Q, the addition to capital stock as a result of the current account must be in surplus or investment equation (6). CAB >0.

    ΔQ = MPK (I)

    The current account balance is accordingly in surplus when aggregate domestic I1

    expenditure is less than output – any point or, = ΔQ MPK

    below the 45° line. The current account balance will by similar reasoning be in Initial levels of investment are assocideficit when aggregate domestic expendi-ated with high marginal productivity and ture is greater than output – any point as investment increases the marginal proabove the 450 line. We now draw in the ductivity of capital diminishes. With inconsumption function (2) in Figure 1 vestment I being measured on the vertical which is a linear function of national axis and the change in output on the output. Also, as not all income (output) is I

    horizontal axis, is initially relatively

    consumed, c1 < 1, and the slope of the ΔQ consumption function is less steep than flat as the marginal productivity of capital

    is high – (1/MPK) low – and becomes steeper as I and ΔQ increase. Thus, the investment opportunities schedule which begins at point A in the diagram, is initially somewhat flat, and becomes steeper at higher levels of output – the dashed line from A. Unlike in short-run models where investment has only an aggregate demand effect, here, the additional capital stock also has an effect on the aggregate supply side where it raises the national output. On the demand side how much investment expenditure will result? Investment should take place up to the point where the benefit from that investment equals the cost. The benefit from investment is the additional output it makes possible or the marginal productivity of capital, and the cost is the interest rate. Hence, investment occurs up to the point where

    MPK = r* = r …(11)

    The slope of the tangent to the investment opportunity schedule at point D has slope 1/r* = 1/MPK and determines the investment expenditure as distance DE = distance BA in the diagram. Distance FA represents the amount of national savings St that finances investment. The remainder of investment (distance BF) is financed by the savings of foreigners and represents the current account deficit CABt. Thus, as equation (8) makes clear, investment is financed by national savings and by running a current account deficit. The demand for foreign currency to finance the current account deficit is met by the supply of currency through capital inflows resulting in a capital account surplus KAS as given in the diagram.

    The slope of the tangent to the investment opportunity schedule at point D is r*CADt. The current account deficit was financed by incurring financial liabilities to the rest of the world – equation (4) – and those liabilities7equal to F are serviced at the interest rate r* in the subsequent period. The distance JK then is the interest payments on the foreign liabilities held r*F and must be subtracted out of national output Qt+1 to arrive at the national income Ynt+1– see equation (3). Unlike in short-run models where capacity is given, here, output in period t+1 is above the output of period t due to the additional investment that has augmented the capacity of the economy to produce. Also, the diagram demonstrates that the resort to foreign savings as a source of financing investment can be maintained as long as

    Economic and Political Weekly November 4, 2006

    Figure 2: Effect of Reduction in Fiscal Deficit opportunities are sufficiently productive


    the servicing of the debt is smaller than the increase in output produced from the investment undertaken. External deficits facilitate higher investment and growth of output so long as the return on the investment funded by this means is beyond the cost of servicing the debt incurred.

    What is the link between fiscal deficits and external sector deficits? Suppose the government reduces its expenditure from G to G/ . Then, public savings rise as Sgovt = T – G and so do national savings S = Yn – C – G. This reduction in the fiscal deficit results in a downward shift of the


    C(Yn) + G schedule to C()Y +G/ in


    Figure 2. The increase in savings to St reduces the expenditure in the current period as well as the current account deficit (from BF to B/F). However, the higher national saving increases the sources of domestic finance for investment whilst the reduced current account from BF to B/F results in a decline in the servicing of external debt. The higher national saving and the lower external debt servicing together result in a higher national income



    in subsequent periods, Y Y t+1 . Thus

    t+1 >

    unlike in short-run models fiscal contractions can actually be expansionary in the intermediate run as capacity gets augmented. The mid-1985 disinflation programme in Israel which included severe fiscal and monetary restraint with the public sector domestic deficit falling to 2 per cent of GDP from 12 per cent prior to stabilisation and being accompanied by


    output booms is a good example of this. Similarly in Denmark in the early 1980s and Ireland in the late 1980s large budget cuts following a stabilisation programme improved the current account and national income.

    There are certain nuances, of course, that the exposition above does not address. For instance, if fiscal contraction is accompanied by a reduction in public investment and infrastructure spending instead of public consumption expenditure, then it will diminish the output generating capacity of the economy and will result in lower output and income in the future. The composition of deficit reduction is also as important as the size of the reduction. Second, the reduction in government expenditure must be perceived to be permanent rather than temporary. If agents in the economy perceive the fiscal contraction to be temporary then agents do not expect a change in future tax obligations required to service the public debt and so private saving and consumption will not respond, thereby reducing the growth promoting impact of a fiscal deficit reduction.

    In an intertemporal set up then lowering fiscal deficits can be expansionary. This is not to deny the possibility that economies with low savings can sustain large current account deficits and yet can experience good economic growth unlike the position of the approach paper to the Eleventh Plan. If the investment that they result in an output that is capable of servicing the increased foreign indebtedness there is no reason for foreign investors to be reluctant about channelling their funds and savings into such an economy. Economies with low national savings can thus undergo periods of high output growth. It is the productivity of investment as depicted in the slope of the investment opportunity schedule that is often a key determinant of sustained economic growth. From an economic policy point of view supply-side measures that improve productivity growth8 can be much more important to growth than trying to spend our way out of current problems.




    1 In this piece we are not going into why large deficits emerged. On that see for one view Joshi and Little (1996).

    2 See Buiter and Patel (2006) and D’Souza (2005). We do not subscribe to the Buiter-Patel intertemporal position that the debt should revert to its original level to be sustainable. That it is converging is sufficient for sustainability for us.

    3 This point is made by W Buiter and U Patel (2006).

    4 This assumption is also made in Mundell-Fleming and intertemporal two-period current account models. The assumption of perfect capital mobility can be relaxed but it would not affect the main results.

    5 When F = 0, NFI = r*F = 0, and CAB = (X – M).

    6 A current account deficit must be financed by capital inflows which is the savings of foreigners deployed in the domestic market.

    7 In this case as there is no initial stock of foreign assets ΔF = F – 0 = F.

    8 On the focus of productivity enhancing policies that are employment promoting as well see E D’Souza (2005).


    Buiter, W H and U R Patel (2006): ‘Excessive Budget Deficits, a Government-Abused Financial System, and Fiscal Rules’ in S Bery, B Bosworth and A Panagariya (eds), India Policy Forum 2005-06, Volume 2, Sage Publications.

    D’Souza, E (2005): ‘A Leap in the Dark: The Fiscal-Monetary Nexus’, H T Parekh Finance Forum, Economic and Political Weekly, Vol XL, No 15, pp 1488-89.

    – (2005): ‘Hitching Employment to a Growth Strategy: Services Bulge and Productivity Growth’, Economic and Political Weekly, Vol XL, No 48, pp 5114-19.

    Joshi, V and I M D Little (1996): India’s Economic Reforms, Oxford University Press.

    Economic and Political Weekly November 4, 2006

    Dear reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here


    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top