Nawaz A. Hakro

8/7/2019 Nawaz A. Hakro

1/37

The Lahore Journal of Economics11 : 1 (Summer 2006) pp. 35-62

IMF Stabilization Programs, Policy Conductand Macroeconomic Outcomes: A Case Study

of Pakistan

Nawaz A. Hakro* and Wadho Waqar Ahmed**

Abstract

This study is designed to assess themacroeconomic performance of fund-supported programs, and the sequencing and ordering ofmacroeconomic policies in the context of the Pakistaneconomy. The generalized evaluation estimatortechnique has been used to assess the macroeconomicimpacts of the IMF supported programs. GDP growth,inflation rate, current account balance, fiscal balance andunemployment are used as the target variables in orderto gauge economic performance during the programyears. The vector of policy variables (that might havebeen adopted in the absence of programs) and the vectorof foreign exogenous variables are also taken asexplanatory variables in the model, so that the individualeffect of the IMF supported programs could be assessed.The result suggests that as the IMF prescriptions wereapplied, the current account balance has worsened, theunemployment rate has significantly increased, and theinflation rate has increased during the years of fund-supported programs. Only the budget balance has shownsigns of improvement. Furthermore an inadequatesequencing of reforms has contributed to the furtherworsening of the economic scenario during the programperiod.

* Dr. Ahmed Nawaz Hakro is Assistant Professor Economics, Quaid-e-AzamUniversity, Islamabad.** Management Officer, Bank Alfalah Limited.


2/37

Nawaz A. Hakro and Wadho Waqar Ahmed

Introduction

Stabilization policy can be defined as the policyresponse to correct macroeconomic imbalances when aneconomy is off track from its potential growth. The generalgoals of stabilization policy are: a) stable growth rate b)stable price level and c) high level of employment (lowunemployment). There are no conflicts over the goals ofthe stabilization policy but conflicts arise over the waysthese objectives are achieved. The International MonetaryFund provides funds to the member country when it faces

balance of payments problems that cause severemacroeconomic disequilibrium in the economy. Besides, asprovided in Article V of the Articles of Agreement of theIMF, it can also impose adequate safeguards. Theseadequate safeguards take the form of policy packagesas conditionalities to the loan. In their practical applicationover time, these policies produced a three-prongedapproach to confront balance of payments problems: (i)securing sustainable external finance (ii) adoption ofdemand-restraining measures and (iii) implementation ofstructural reforms. IMF adjustment programs are of two

orientations: a) short-term, in which the macroeconomicdisequilibrium is thought to be reversible in one or twoyears, and b) medium-term in which the macroeconomicdisequilibrium is caused by structural impediments togrowth or a heavy external debt burden. The StandbyArrangement (SBA) is an example of the IMF short-termprogram. The priority course of action in SBAs isexpenditure reduction. IMF medium-term programs aim tocorrect a serious external payments disequilibrium due tostructural impediments to growth and debt overhang. Theprogram involves a strategy that keeps expenditures in

line with output and increases growth. Examples of theseprograms are the Structural Adjustment Facility (SAF),Extended Structural Adjustment Facility (ESAF) andPoverty Reduction and Growth Facility (PRGF).

36


3/37

IMF Stabilization Programs, Policy Conduct and MacroeconomicOutcomes

Pakistan accepted fund supported adjustmentprograms in the 1980s1 and has become a prolongedborrower with more than 15 years of borrowing withAdequate Safeguards. We do not find many studiesanalyzing the impacts of these Adequate Safeguards onthe Pakistan economy. Yet there are a few; as Kemal (1994)has shown the employment situation further worsened dueto privatization, and structural adjustment has beenaccompanied with rising inequalities and poverty.Stabilization and growth are not mutually exclusive andany policy has to incorporate both elements. However, the

manner in which the policy has been implemented inPakistan has tended to pursue stabilization at the expenseof growth is the conclusion of Bengali and Ahmed (2001).Pakistan has been unable to sustain high economic growthwith equally impressive reductions in poverty (Khan, 2002).Real output declined, the inflation rate increased, and theexports of goods remained insignificant during theadjustment period 1988 to 1991 but the findings show thatadjustment lending enhanced investment and increasedthe governments current consumption (Iqbal, 1994).

Consequently, this paper is an attempt to analyze themacroeconomic outcomes of the fund- supported programsin the context of the Pakistan economy. The generalizedevaluation estimator technique has been used, which isconsidered to be a better technique than the available onesin the literature of the fund-supported programs (seeMethodology), and is a first attempt of its kind in Pakistan.Along with the macroeconomic outcomes due considerationhas been given to the sequencing and ordering of the policyconduct that is very important to assure the effectiveness ofany program. The success of the programs is measured in

terms of the macroeconomic outcomes, but it is not astraightforward task to define the effectiveness of theadjustment programs. It is very easy to check theimplementation of those policy changes on which both

1 Pakistan has had a long association with the IMF; it joined the IMF on July11, 1950. Thefirst time when the Government of Pakistan asked for a loan was 1958. As the IMFsfunding amount and pattern changed after the 1970s, right after a couple of shocks of oil

price and debt crises of 1980s, it was 1988 when Pakistan accepted policy packagessuggested by the IMF.

37


4/37


parties (IMF and country) are mutually agreed upon. But it ismuch more difficult to know whether these changes will leadto the desired macroeconomic targets at least for tworeasons. First, from a theoretical point of view, the fundsupported program is the composite of the complex policypackages that include monetary and exchange rate policies,fiscal measures, policies to improve efficiency, tradeliberalization, price and wage reforms, privatization andfinancial sector reforms. The theory underlying the dynamiclinkages among such policy packages combining demandmanagement policies with supply enhancing policies and a

set of multiple macroeconomic targets is not wellestablished. As Baqir, Ramcharan and Sahay (2003) foundby regressing the deviation between the programmed andactual growth on the deviation between programmed andthe actual values of the current account, better performancein the current account is accompanied by worseperformance in terms of growth. Second, the fund-supportedprogram is only one of the exogenous shocks that hit theeconomy of the typical country. Other external shocksinclude changes in the terms of trade, changes in the cost ofdebt servicing, droughts, famines etc. The Afghan crisis, the

incidence of September Eleven, nuclear tests of 1998 andthe Iraq war have greater implications on the Pakistaneconomy.

The paper is organized as Section-I Introduction;Section-II Literature Review; Section-III Methodology andmodel, Section-IV Results and discussion, Section-VSummary and conclusion.

Literature Review

Being an old addict of fund-supported programs,the Pakistan economy presents a very good case studyfor analyzing the impact of fund-supported programs.There is not much literature available for the assessmentof the fund-supported program of a typical economy.Quite a few studies are available but they also need to bemodified in terms of the assessment of themacroeconomic outcomes. There is even more of a lack

38


5/37


of literature about the sequencing and ordering of thepolicy reforms. Certainly this would be a novelty to bringthe sequencing and ordering into the picture whileassessing the outcomes of the fund-supported programsand analyzing the macroeconomic impacts of theprograms on the Pakistan economy.

Economists of different schools of thought haveanalyzed the stabilization policy and structural adjustmentprograms and their impacts on the balance of payments indifferent ways. Conflict over the results of these fund-

supported programs on the macroeconomic variables,especially on balance of payments is there from the verybeginning of these programs. Few studies found that thesepolicy reforms work in terms of improving the balance ofpayments position. As Bagci and Perraudin (1997) foundby using a generalized evaluation technique, fund-supported adjustment programs improve the overallbalance of payments performance of the countriesinvolved in the programs. Schadler et.al. (1993) get thesame results using the before after technique and Khanand Knight (1981), (1985) obtain similar conclusions. On

the other hand Loxeley (1984), Connors (1979), and Moran(1989) have found that the fund-supported programs haveno significant impact on the balance of payments positionof those economies that accepted the fund-supportedadjustment programs.

Pakistan, like other developing countries,experienced balance of payments deficits throughout the1980s. Iqbal (1994) finds that in the case of Pakistanoutput declined due to fund supported programs, theinflation rate increased and exports remained insignificant.

The evaluation of the three-year program shows that theapplication of the short-term policies to the long-termadjustment problems resulted in a number of policyconflicts. The factor behind the sharp acceleration of thecredit and money supply shows that the policy of creditrestraint is in direct conflict with the objective of pricestabilization in a less developed economy, even when itsoutput constraint is overcome in any period. Similarly, the

39


6/37


objective of improving the balance of payments byencouraging exports remained unfulfilled, and theinflationary situation has worsened, Bilquees (1987).

Another important macroeconomic target underconsideration is economic growth. Economic growth is avery crucial macroeconomic variable to gauge theeconomic performance of any country. Do fund supportedprograms lead to improvement and sustainability ineconomic growth? This has been a central question indiscussing the performance of fund-supported programs.

Kiguel and Livatian (1992) find that the programs that usethe exchange rate as a main nominal anchor are oftenassociated with a business cycle that begins with boomand ends with recession. While the programs that usemoney supply as the main nominal anchor generallyinduce the usual Philips curve result, lower inflation isaccompanied by recession after the program isimplemented. Bruno (1992) concludes that recentexperiences of IMF-supported programs in Hungary,Poland, Czechoslovakia, Bulgaria and Romania ended withoutput collapse. Cukierman and Liviatan (1992) show that

when the difference in the ability of strong and weakpolicy makers to control inflation is large, unexpectedinflation may be persistently negative for quite a while,thus causing reduced economic activity and indicating thatcredibility is low. Uribe (1999) found that exchange ratebased (ERB) and money based with initial reliquefication(MBR) programs induce an initial expansion in theeconomy while the money based (MB) programs areinitially contractionary. Balassa (1982) concludes thatcountries that applied an outward oriented strategy hadfavorable growth experience after 1973. If the

government adopts public sector price increase combinedwith tough layoff policy, there is a strong presumption thatreal output will be higher and inflation would be lower inthe overall time horizon, Buffie (1992). The conclusionfrom Khan and Knight (1981) is that programs designed toachieve quick results on the balance of payments viasharp deflation are likely to have significant andundesirable impacts on output and employment,

40


7/37


particularly in the short run. Ball and Sheridan (2003) haveconcluded that there is no evidence that inflation targetingimproves the performance of the economy with regard tooutput growth. Hutchison (2001), using 461 IMF-stabilization programs and 160 currency crises, found thatcurrency crises, even after controlling for macroeconomicdevelopments, political and regional factors significantlyreduced output by 1 to 2 percentage points. Loxley (1984)and Connors (1979) found contractionary effects onoutput. Baqir et. al. (2003) have reported results fromregressing the deviation between programmed and actual

growth on the deviation between programmed and actualvalues of the other program objectives that there is anegative and statistically significant relationship betweengrowth and current account objectives. Stiglitz (2000)concludes that IMF economic remedies often make thingsworse, turning slowdowns into recessions and recessionsinto depressions.

The Pakistan economy unlike other developingcountries enjoyed healthy economic growth during the80s, averaging above 5% per annum. But after the

adoption of fund-supported programs, economic growthstarted to decline rapidly. As Bengali and Ahmad (2001)have concluded Stabilization and growth are not mutuallyexclusive and any policy has to incorporate both theelements. However, the manner in which the policy hasbeen implemented in Pakistan has tended to pursuestabilization at the expense of growth. It has dampenedinvestment and curtailed purchasing power, leading to arecessionary situation. It has contributed directly to theincrease in unemployment and poverty. Monetary growthin Pakistan is to some extent anticipated. There was no

evidence that only unanticipated policy has a real outputeffect, as discussed by Khilji and Leon (1989). Growthcontributes more to poverty reduction when it increasesemployment, productivity and the wages of poor people,and when public resources are spent on humandevelopment and physical infrastructure. Pakistan hasbeen unable to sustain high economic growth with equally

41


8/37


impressive reductions in poverty, as discussed by Khan(2002).

After the shocks of the 1970s, most developingcountries were running fiscal deficits, and their eradicationtook the focus of attention in adjustment programs.Excess of expenditure over revenues is considered to bethe cause of worsening balance of payments deficits andinflation. Bulir and Moon (2003) show that the overall fiscalbalance of sampled countries improved in most cases inthe 1990s but the impact of IMF supported programs was

not statistically significant. Franco (1990) also found thatbalance of the budget itself is not sufficient to establishthat these reforms effectively took place, since inflationaffects the budget deficits in various respects, so theinfluence of price stability on deficits might very well be animportant part of the explanation of these sudden budgetimprovements. In the context of the Pakistan economy,Ahmad (1998) has found that the experience ofimplementing fund supported reforms reveals that non-fiscal policies have mostly conflicted with fiscal policies inachieving fiscal discipline. At the cost of painful tradeoffs,

the fiscal deficit has come down from 7% of GDP in the1980s to 5.4% of GDP in 1997-98. This reduction in thefiscal deficit seems to have slowed down the growthtempo, which in turn has reduced revenue potentials.

Sequencing of reforms is the order in which eithermacroeconomic policy actions or specific reforms areintroduced. It involves the order in which reforms areundertaken across the different sectors. A distinction insequences of different reforms is needed among theeconomies because of the initial differences. In developing

countries fiscal, institutional and monetary reforms shouldbe taken first rather than trade reforms, financial reformsand capital account liberalization (Nsouli, Rached andFunke (2002)). Ivanova (2003) concludes thatimplementation of the program primarily depends uponthe borrowing countrys political economy. Politicalinstability, strong interest groups, inefficientbureaucracies, lack of political cohesion, and ethno-

42


9/37


10/37


programs, such as higher growth and lower inflation, havenot manifested themselves in Pakistan, with growthconsiderably low and inflation high. Kemal (1994) hasshown that despite containing the employment costthrough limiting the wage rate and reducing employmentby about 15%, the non development expenditure andfiscal deficit have continued to increase, the employmentsituation has further worsened due to privatization, andstructural adjustment has been accompanied with risinginequalities and poverty in the Pakistan economy. The Ginicoefficient increased from 0.34 to 0.41 and the proportion

of the poor has increased from 13% in 1987-88 to 14% in1990-91. Amjad (2004) has concluded that the disaster interms of economic decision-making during the reformperiod in Pakistan characterized the financial reformprogram adopted in the late 1980s. By drastically raisingthe interest rates to market prices on governmentborrowing, it increased many fold the interest paymentburden of the government.

Methodology

In the literature, five types of approaches for theassessment of the macroeconomic outcomes of fund-supported programs are found as very common:

The before-after approach

It compares the macroeconomic performance of aneconomy before and after the adoption of fund supportedprograms2.

The with-without approach

2 This is the most popular approach in the early literature of the fund-supported program.The first study to use this was by Riechman and Stillson (1978). It compares themacroeconomic performance before and after the adoption of a fund-supported program,assuming all other things constant. It has the advantage of ease of calculation but wheneverthe other factors that are assumed constant by this approach affect the macroeconomic

position of the economy, it fails. This is because the fund-supported program is one of thoseexogenous shocks that hit the economic variables. So if there are other factors like terms oftrade, industrial growth, movements in the interest rates etc., along with the fund-supported

program that affects the economic variables of the country, it produces biased results.

44


11/37


It compares the macroeconomic outcomes of fund-supported programs by differentiating the program andnon-program countries3.

The comparison of simulation approach

It relies on the simulations of econometric models toinfer the hypotheticalperformance of the policies includedin fund supported programs and alternative policypackages4.

Actual versus targeted approach

It compares actual outcomes for certain keymacroeconomic variables to their respective targets, forsuch variables specified by the authorities and the fund atthe inception of the program5.

The generalized evaluation estimator approach

3 This is another counterfactual approach, which tried to overcome the drawback of thebefore-after approach. As this takes the panel of the program and non-program countries andassumes that both the program and non-program countries have the same non-program

determinants. Ball and Sheridan (2003) and Fisher (1988) use this approach. Though itovercomes the problem of the before-after approach, it also has a few inherent problems.The assumption of the same non-program determinants between the program and non-

program countries is quite unrealistic. Because the countries selected are not takenrandomly, in fact they are selected for having a poor economic performance prior to the

program. This implies that the program countries had a weaker position prior to the programthan non-program countries. So macroeconomic determinants between these two groups ofcountries would not be the same, in this position the with-without approach would produce

biased results.4 This approach differs from the other three in that it does not consider the actualoutcomes of the program but it relies on the econometric model to incorporate theimpacts of the fund-supported program. Khan and Knight (1981) and Khan and Knight(1985) have gauged the macroeconomic impacts of the fund-supported program. Butthis approach carries a famous Lucas critique that the actual effects can turn out quite

different from the simulated ones. Second, due to credibility factors, the effect of thepolicy can be different when it is implemented inside and outside the fund-supportedprogram.5 This is not one of those approaches that are frequently used in the literature of the

fund-supported programs. It compares the targets set by the programs for certainvariables of interest to the actual outcomes of the program. The drawback of this is thatthe targets are mostly not available to the public. It is also deficient because targets may

be overly ambitious so that failure to achieve them does not necessarily imply that theadjustment program was not effective. Conversely, the targets could be under ambitiousso that exceeding them does not always mean that the program was effective.

45


12/37


It compares the macroeconomic performance of theprogram and non-program countries, adjusting for theinitial differences and condition among the countries andcontrolling for exogenous influences.

The recognition of the inherent biases in with-without and before-after approaches led to the creationof the generalized evaluation estimator approach. Itmodifies the with-without approach in two ways. First, itaccepts the non-random selection of the countries, andidentifies the specific differences between the program

and non-program countries in the pre-program period.Second, it attempts to capture the effect of policy,exogenous shocks and other variables on themacroeconomic outcomes, taking into account howpolicies would have evolved in the absence of theprogram. The reaction function captures the effect of thepolicy. The reaction function brings those policies intoconsideration that might have been adopted in theabsence of the program. The reasoning behind thecounterfactual approaches is that either they comparethe before and after situation in the economy or

compare the sample of the program countries with thesample of those countries that are not involved in theprogram.

At the same time the country cannot be in bothsituations (program and non-program). Therefore, thereaction function is estimated by taking the difference ofthe vector of desired values of target variables with thevector of actual values of the target variables in the lastperiod. Second, by taking the exogenous shock as anexplanatory variable in the model, it provides a good

measure for analyzing their impact on the target variables.While applying it to the case study it also overcomes thetwo well known limitations, selection of random countries,and the problem of the degree of programimplementation, i.e. the willingness of governments toimplement certain programs. Second, technically themethodology used in this study overcomes the problemsassociated with the with-without, before after and the

46


13/37


other above discussed techniques, so the IMF estimatorestimated with this technique would not provide biasedresults. That is why it is known as the comparatively bettertechnique to evaluate the fund-supported programs.

This technique is capable of serving the objectivesset by this study. First by taking economic growth, currentaccount balance, inflation rate, budget deficit, andunemployment as the target variables, we can estimatethe impact of the fund-supported programs on themacroeconomic scenario of the Pakistan economy.

Second, by including the reaction function in the model italso deals with the difference between the targeted andactual outcomes and it concentrates on the effect of theother policy options that are not included in the programdesign by taking the vector of the policies which mighthave been adopted in the absence of the programs. Third,including the vector for the exogenous variables, theireffect on the target variable is also incorporated.

The Model

Suppose that the target variable is determinedaccording to:

Yi = 1 + 2xi + 3 wi + IMF d + (1)

Where Yi is the target variable (i = Current accountbalance, economic growth, inflation rate, fiscal deficit,unemployment and foreign exchange reservesrespectively), xi is a vector of policy instruments (i.e. theexchange rate, fiscal deficit, domestic credit, inflationrate), w are foreign exogenous variables (eg. Terms oftrade, international interest rate), d is a dummy variable

and is a random shock. The dummy variable takes on thevalue 1 if the fund-supported program is in effect duringthe period in question and zero otherwise. The parameterIMF measures the effect of the program during this periodon variable y.

It is important to note that the definition of meansthexi refers to the policies that would have been adopted

47


14/37


in the absence of a program. The vector xi is thereforedirectly observable only if there is a fund-supportedprogram; for non-program ximust be estimated. One wayin which xi can be estimated is via the simple reactionfunction:

xi = [yid- (yi)-1]+ (2)

Where yi is a vector of target variables, yid is the vector oftheir desired values, is an adjustment parameter, i is a

vector of random shocks, and is first-difference operator.Equation (2) basically says that the change in thecountrys macroeconomic policy instruments between thecurrent and previous period will be the function of thedifference between desired values of the target variablesthis period and their actual values last period.

The model can be employed to examine thestatistical properties of the beforeafter and with-withoutapproaches to an estimation of the program effects. Thebefore-after estimator BA is:

BA = yi for i P (3)

Where P denotes the set of program countries during thecurrent period. The expected value of this estimatorconditional on observed values of the foreign exogenousvariables is:

(BA\ iP,w) = IMF+3 w+(2xi+\ iP,w) (4)

Which is equal to the true value ofIMF only if:

(2xi+\ iP, w) = - 3w(5)

The before-after estimator is unbiased if one expectsthat the non-program determinants of yi would havebehaved in a way to leave the y i unchanged, on average,between the program and the non-program periods. Inother words any change in the external market, innovation

48


15/37


in policies, and other unobserved variables cancel eachother out.

The with-without estimator ww is given by:

ww = yij - ynj (6)

Where ynj is the average value of the yij over some set Nof non-program countries. Since we can observe x and ijfor all iN, the information set, defined as , now consistsof:

= {(xi, ij for iN), w}

Taking expectations ofww conditional on iP and we have:

(ww\iP,)=IMF+(2 xi+ij\iP, ) ( 2 xn +nj )(7)

The with-without estimator will be unbiased if:

(2xi +ij \ iP,) = 2 xn + nj (8)

In other words, if it can be expected that in theabsence of the program, the country would have behaved just like the average member of the non-programreference group, then the estimator will be unbiased.

An alternative to the before-after and with-withoutapproach can be derived by using equation (2) tosubstitute out the unobservable policy changes that wouldoccur in the absence of a fund program (i.e. for xi) fromequation (1). The generalized evaluation estimator is:

yi = oi (yi)-1 (2 + 1) + (xi)-12+3 w+IMF d+( +2) (9)

Where yid is subsumed into the constant such that oi = 1+ 2yid.

Econometric estimation of equation (9) produces anestimate of the IMF that is not subject to the criticism

49


16/37


leveled at the before-after and the with-withoutestimators. This equation takes care of estimation of thecounterfactual by controlling for the factors that aresystematically related to the policies that would have beenfollowed in the country without the program, which is toinclude the lag values of the target variables and thepolicy instruments in the specification. The equation wouldbe estimated by OLS estimation.

Results and discussion

The model is nested as to test the simultaneouseffect of, a) the IMF programs, b) policy shocks, and c)foreign exogenous variables. It serves our objective ofdiagnosing the impact of fund-supported programs, whiletaking into account the effect of other policy options thatmight have been adopted in the absence of the programsand foreign exogenous variables. The model takes the IMFprograms, other policy responses and the foreignexogenous variables as different explanatory variables,which makes it easier to gauge the net effect of the fund-supported programs on the target variables. The targeted

macroeconomic variables are the current account balance,unemployment, GDP growth, inflation rate and thebudgetary balance6. The annual data from World Bankdata sources for Pakistan have been taken from 1973 to2000.

Equation (9) is estimated for all five-target variables,discussed above. As we deal with time series data, theAugmented Dickey-Fuller Unit Root Test is used to check

6 The selection of the target variables to gauge the macroeconomic performance of thetypical fund supported program is very crucial, in this study five target variables aretaken, namely: GDP growth, Inflation Rate, Unemployment, Fiscal Deficit, and CurrentAccount Deficit. The rationale behind the target selection is very simple as discussed inthe introduction earlier that the general goals of any stabilization policy are; stablegrowth, low and stable rate of inflation and high level of employment (lowerunemployment). For this reason GDP growth, inflation rate and unemployment have

been taken as the target variables so as to see how much these programs are successfulin attaining their basic goals. The rationale behind the selection of the current account

balances and fiscal deficit is that the prime responsibility of the IMF is to assist themember country when it faces a balance of payments problem and according to theirapproach, fiscal imbalance is the main culprit behind the balance of payments problem.

50


17/37


for stationarity. The results suggested that the data isstationary for all the target variables. The statisticalparameters for the overall significance are represented inTable-1. The R2 is quite high for all equations. That is .85, .79, .77, .64 and .74 for the equations of the currentaccount balance, unemployment, GDP growth, budgetarybalance and the inflation rate, respectively. These resultsshow that in all equations except the budgetary balancemore than 75 percent of variation in target variables isdefined by the explanatory variables. The F statistics are18.7, 12.27, 13.9, 7.3 and 11.7 for all equations of current

account balance, unemployment, GDP growth, budgetarybalance and the inflation rate, respectively. It is highlysignificant for all equations, which clearly tells us that allparameters of the explanatory variables are non-zero. TheQ statistic is used for the detection of autocorrelation orpartial autocorrelation. The final results find no auto orpartial autocorrelation, while the White-Hetroskedasticitytest shows no evidence of of hetroskedasticity.

(1) The current account balance

The regression results for the current accountbalance show that the parameter of the IMF dummyindicates a negative impact on the current accountbalance, which is statistically significant at 1 percent levelof significance, and the t statistic is -2.67, which is quitehigh. The inflation rate (CPI) policy variable has alsoworsened the current account variable and the result issignificant at the 5 percent level. Another policy variable,net capital account (NCA) has shown positive significance,and indicates that the net capital account has caused animprovement in the current account balance. Though its

parameter is significant at the 5 percent level ofsignificance, the magnitude of the parameter is very small. The structural dummy has been taken as the foreignexogenous variable, which has also a positive significantimpact over the current account position at the 5 percentlevel of significance. The parameter of current accountsown lag taken as an explanatory variable shows that thecurrent account is positively related to its own lag. The

51


18/37


average effect of all the other variables that are notincluded in the model (represented by constant C) isnegative.

The results indicate that due to IMF supportedprograms Pakistans current account situation has furtherdeteriorated. Continuous attempts to correct balance ofpayments imbalances through liberalization of theeconomy probably leading to an immediate andundifferentiated reduction in import tariffs, which has notgiven national industries adequate time to improve their

competitiveness with foreign firms. The other reason isthat the Pakistan economy does not have a diversifiedportfolio of exports, and it mostly relies on the export oftextile family products and a few agricultural products.Continuous depreciation of the rupee has not shown anypositive signs in terms of expanding the demand forPakistani exports and has not shown any satisfactoryresults in terms of diminishing the demand for foreigngoods that are quite price inelastic in Pakistan, becausemost of Pakistans imports consist of capital goods that areused as an input in domestic industry.

The other policy variable, net capital account (NCA)plays an accommodating role in terms of compensatingthe current account deficits and leaving the balance ofpayments in better condition. At the same time, theinflation rate (CPI) has worsened the current accountbalance situation in Pakistan. Though it has notsignificantly increased the prices of exports, the primecause is that during the entire last decade the Pakistaneconomy has experienced imported inflation in terms ofincreases in the prices of imports due to devaluation. The

foreign exogenous variable that is taken as the dummy forthe structural variable, which is positive and significantmight have directly affected the current account balance.

(2) Unemployment:

As presented in Table 1 the regression resultsindicate a very low level of significance and the parameter

52


19/37


is as high as 1.8 that the IMF supported programs inPakistan have worsened the unemployment situation inthe country. In other words the IMF has a positive impactover the rate of unemployment. The policy variable totalexpenditure as percentage of GDP (EXPTGDP) has asignificant effect at the 1% level of significance onunemployment. The sign of the parameter is negativewhich shows that an increase in the total expenditurereduces the unemployment rate. The other policy variableinflation also has a negative coefficient, which implies thatthe inflationary pressures have helped to lower the

unemployment level. The foreign exogenous variableterms of trade (TOT) has shown a negative impact onunemployment which is significant at the 10 percent levelof significance. The results further show thatunemployment is positively affected by one period laggedunemployment that is significant at the 1 percent level ofsignificance. And the average effect of all the othervariables that are not included in the model is also positiveat the 1 percent level of significance.

IMF supported programs have worsened the

unemployment situation in the economy, which was 1.7%of the total labor force in 1970 and has worsened to 7.8%of the total labor force in 2000. Reduction in publicexpenditure is one of the main conditionalities of the IMF inall these programs. The reduction in public expenditure canbe achieved either by restricting the acquisition of thecommodities or limiting the employment cost throughreduction in employment or limiting the increase in thenominal wages below the inflation rate. The decline in theemployment cost has been brought about by containing theincrease in the nominal wages of government employees

and even imposing a complete ban on recruitments andencouraging early retirement. Privatization has alsoinversely affected the level of employment; the new ownershave laid off workers employed in the public sector. Thecontinuous contractionary policies have caused littleexpansion of the economy which has been unable toemploy the growing labor force in the country. During theprograms the Pakistani economy witnessed not only

53


20/37


complete bans on recruitment but also schemes such asthe golden hand shake that were introduced toencourage early retirement and on the other hand acontinuous decrease in development expenditures has alsocaused worsened the employment situation in the country.

The policy variables that might have been adoptedin the absence of the program have a significant negativeeffect on unemployment. Increases in expenditure expandthe productive activities in the economy that provideopportunities for employment and hence reduce

unemployment. As it does not show much increase afterthe involvement in the fund-supported programs, it did notplay a compensatory role in reducing unemployment. Butsurely any expansion reveals an increase in the level ofemployment. The price level shows the well known Phillipscurve relationship in the case of Pakistan. Any increase inthe price level increases the profitability of the industrythat causes its further expansion, or inflation is caused byany expansionary policy and both have the sameimplication in terms of raising the demand for labor andthereby reducing the rate of unemployment. Though the

inflation rate was in double digits almost in all the years ofIMF programs since 1988, except for the last three years,this was not sufficient to compensate for theunemployment created by introducing IMF supportedprograms. The foreign exogenous variable terms of trade(TOT) have a negative effect on unemployment andcontinuous deterioration in the terms of trade has causeda substantial increase in the number of unemployedpeople.

(3) GDP growth:

The results of GDP growth indicate that the IMFprograms have lowered economic growth. Its coefficienthas a negative sign, the t statistic is greater than 1, butthe results are not statistically significant. The policyvariables exchange rate and the inflation rate both havenegative parameters and both are significant at the 5percent level of significance. So the results show that

54


21/37


these are the policy shocks that might have a negativesignificant effect over GDP growth. The foreignexogenous variables have no significant effect overgrowth. It is positively influenced by its own lag, whoseeffect is statically significant at the 1 percent level ofsignificance. The average effect of all the other variablesnot included in the model is positive and significant at the1 percent level.

The regression results point out that the IMFprograms may have slowed down the pace of economic

growth that the Pakistan economy was enjoying before theadoption of the fund-supported programs. That is the maincritique of the fund-supported programs-that thecontractionary policy to diminish the budget deficits andstabilize the prices have been achieved at the expense ofgrowth. But the negative effect of the IMF programs overthe Pakistan economy is not statistically significant. Thereare other policy variables that are responsible for theslowing down of the growth rate of GDP. One is exchangerate devaluation of the Pak rupee which has made importsmore expensive, and since major imports are used as an

input in the domestic industry, devaluation has loweredthe productive activities in the economy. It has worsenedthe situation both ways, one by diminishing the productiveactivities and second by increasing the cost of productionfor the domestic industries that further caused thedomestic prices to increase. So inflation has alsonegatively hit the economic growth of the country bymaking the domestic products expensive and the demandfor imported goods to rise. The overall effect that thePakistan economy has faced in terms of GDP growth afterthe involvement in the fund-supported programs may

possibly be a contraction in the growth rate.

Table-1: Results

CA UE GDP CPI BBDIMF -0.94 1.81 -0.54 5.92 1.01

(-

2.67)* (4.36)* (-1.33) (4.50)*(2.03)*

*

55


22/37


D98 3.07 - - - -

(2.28)*

*

TOT --2.74E-

11 - -5.69E-

11

(-1.86)*** (2.84)*

(ER)-1 - - -0.06 0.74 -

(-

2.02)**(2.50)*

*

(M2)-1 - - - 1.57E-11 -

(2.10)*

*

(BRT)-1 - - - - 1.17

(2.27)*

*

(EXPTGDP)-1 - -0.35 - - -

(-3.72)*

(CPI)-1 -0.14 -0.04 -0.11 -0.46 -

(-

2.50)** (-2.34)**(-

2.09)** (-7.00)*

(NCA)-1

3.48E-09 - - - -

(2.10)*

*

(UE)-1 - -0.23 - - -

(-2.94)*

(GDP)-1 - - -0.72 - -

(-

3.79)*(BB)-1 - - - - -0.29

(-

1.79)***

C -3.15 7.49 6.53 9.62 -13.64

(-3.00)* (3.78)* (3.46)* (3.37)*(-

2.34)**

(CA)-1 -0.79

56


23/37


(-5.00)*R2 0.85 0.79 0.77 0.64 0.74

F statistics 18.70 12.27 13.90 7.30 11.70

Note: (i) The values in brackets are t statistics.

(ii) The *, **, *** shows the 1%, 5% and 10% level ofsignificance, respectively.

(iii) No * indicates insignificant results.

(iv) All the F statistics are significant at very highlevel of significance

(4) Budgetary balance:

The results derived from the budgetary balanceequation suggest that the parameter of the IMF programshas a positive sign and is statistically significant at the 5percent level. As depicted in the table the IMF programshave improved the budgetary balance. They have reducedthe budget deficit, which is a chronic disease for theeconomy. The bank rate policy variable has also a positivecoefficient that is significant at the 5 percent level ofsignificance. So the bank rate has also contributed to thereduction of the budget deficit. The foreign exogenousvariable terms of trade has a positive parameter at 1percent level of significance but it has worsened over thelast twenty years and contributed to the increase in thefiscal deficit. The own lag of the budgetary balance issignificant at the 10 percent level with a negative sign. The constant term has a negative impact over thebudgetary balance, showing the average effect of all theother variables excluding those presented in the modelwhich is negative and significant at the 5 percent level ofsignificance.

This is the only area where the Pakistan economyhas been slightly better off due to fund supportedprograms. The parameter of the IMF dummy shows apositive sign that indicates that due to IMF supportedprograms the fiscal deficit, which was the major imbalancein the economy, has been reduced. The fiscal deficit has

57


24/37


come down from 8 percent of GDP in 1987 to 5 percent ofGDP in 2000. The conditionality posed by the IMF toreduce the public expenditure and increase tax collectionhas contributed to the reduction of the budget deficit.Pakistan has cut the public expenditure from the verybeginning of the programs though the share of taxcollection to GDP does not show much improvement.Second, increasing the autonomous power of the centralbank and financial reforms set by the fund has increasedthe cost of domestic financing of the government. From1991 the government started the full-fledged system of

auctioning of government debt and allowed the rate ofreturn on the treasury bills to rise from the unrealistic 6percent where it was earlier, to a more realistic 13percent. All these financial improvements introducedcompetition in government borrowing from the public, andcaused the increase in the cost of borrowing in terms ofoffering higher rates of return on treasury bills and othergovernment securities.

Bengali and Ahmed (2001) have diagnosed thereduction in the fiscal deficit in terms of the different

distributional implications and this is quite important tonote here. For example, raising revenue or reduction inexpenditure can lower the budget deficit. Revenues can begenerated through direct or indirect taxes: the formerimpacts the rich while the latter impacts the poor. In thesame way expenditure can be contained by reductions inthe current expenditure or reduction in developmentexpenditure. The former impacts on existing employmentwhile the latter impacts on future employment creation.The data highlight that revenue shortfalls, current accountoverruns and cuts in development expenditure are the

norm. That is what is indicated above in the resultsdiscussion of the unemployment estimator, that the IMFprograms have reduced the level of employment in thecountry.

The other policy variable that caused improvementin the budget deficit is the bank rate (BRT). The economiclogic behind the response of the government to the

58


25/37


increase in the interest rate is very basic: that the increasein the domestic interest rate forces the Treasury bill rateto increase in order to increase the domestic debtportfolio. In other words the increase in the rate of interesthas increased the cost of borrowing of the governmentfrom the public.

(5) Consumer Price Index (CPI):

The inflation estimator provides the result that IMFprograms have contributed to the increase in the rate of

inflation in the economy. The very high parameter of theIMF dummy (5.9), with the positive sign and significant at1 percent, shows that IMF programs have brought aconsiderable increase in the rate of inflation. The otherpolicy variables, money supply and the exchange rate, arealso contributors to the increase in the rate of inflation.Money supply has a very small parameter estimatesignificant at the 5 percent level. The exchange rate hasalso a significant parameter at 5 percent. Foreignexogenous variables have no significant impact on theinflation rate. The parameter of its own lag has a negative

sign and significant at 1 percent indicating that it ispositively related to its own lag. The effect of all othervariables not included in the model is positive andsignificant at the 1 percent level of significance for theinflation equation.

The large and positive parameter estimate on thedummy of the IMF implies that the involvement in thefund-supported programs has increased the rate ofinflation in the economy. It was in double digits except forthe last three years. Though most of the policies by thefund-supported programs are contractionary in theirnature, the main policy regarding trade is the depreciationof the currency, because IMF and other financialinstitutions believe that currencies in most developingcountries are over valued so they must be rationalized.After the Bretton Woods system in which the rupee waspegged to the dollar, the Pakistan economy is operatingunder the managed floating exchange rate. The

59


26/37


depreciation of the Pak Rupee increased the prices ofmachinery and crude oil. Both are the basic inputs indomestic industry, and the rise in their prices increasedthe pressure of cost-push inflation, the main source ofinflation in Pakistan. But for the last three years the rate ofinflation has come down due to the continuouscontractionary policies on both counts in terms of fiscalpolicy by reducing public expenditure, and by monetarypolicy by reducing the growth of the money supply. Butprice stability has been achieved at the expense of GDPgrowth, increasing unemployment and increasing poverty.

There is no significant effect of the foreignexogenous variable on the CPI. The other policy variables,the exchange rate and money supply (M2), have alsocontributed to the increase in the rate of inflation in thePakistan economy. The logic for the exchange rateparameter is the same as what has been discussed aboveby forcing cost-push inflation in the economy. Increases inmoney supply contribute to an increase in domesticdemand for goods and causes prices to rise, which isknown as demand-pull inflation. But the coefficient of M2

is so small that its effect can be ignored. So the regressionresults suggest that inflation in Pakistan is caused by theincrease in input prices, or in other words cost-pushinflation is dominant in this case.

The sequencing of reforms:

There has been much discussion regarding thesequencing of the reforms. Results of macroeconomicoutcomes discussed earlier show that the IMF programshave increased the unemployment rate, increased theinflation rate, worsened the current account deficits andcontributed to slowing down the pace of the economy.Besides the other reasons, inadequate sequencing ofeconomic reforms is also a contributor to these results.The first indicator of the implementation of reforms can bethe completeness of the arrangements settled with theIMF, as these arrangements are based on theconditionalities set by the IMF. Inability to meet the

60


27/37


minimum requirements causes the termination ofdisbursement of a loan before the expiration date andcompleting the approved amount. After 1988, Pakistan hasonly once in 2000-01 has been able to draw the fullamount of the approved arrangement. It shows Pakistanhas been unable to meet the target requirements set bythe IMF. That clearly shows inadequate sequencing ofreforms.

First on the agenda of these economic reforms wasmacroeconomic stability consisting of three ingredients:

budgetary balance, balance of payments improvementand reduction in inflation. Fiscal and monetary policyreforms and devaluation have mainly aimed at achievingthis (Ahmad, 1998). The normal sequencing of the fundreforms is first to improve the budgetary surplus and toreduce the price level. In the case of Pakistan, financialsector reforms were adopted in the very beginning of thereform packages that increased the competitiveness ofthe government to generate funds from the public andresulted in an increase in the rate of return on treasurybills and other government securities. This increase in the

interest rate increased debt servicing of the governmentby many fold. Clearly the adoption of financial sectorreforms was not a suitable strategy, which not onlyincreased government expenditure but also reduceddevelopment expenditure, and contributed to the slowingpace of the economy and poverty that was reduced in the1980s then again went up very rapidly.

It is important to restructure state enterprises, andprivatization is often an effective way to do so. But movinga person from low productivity jobs in state enterprises to

unemployment does not increase the countrys income,and it certainly does not increase the welfare of theworkers. Sufficient effort was not made to increaseemployment in order to counteract unemployment createdby privatization and contractionary policies. As Stiglitz(2002) pointed out, macroeconomic stabilization is on theagenda of the IMF supported programs but job creation ismissing. The objective of the reduction in fiscal deficit by

61


28/37


reducing the expenditure posed some serious costs interms of worsening the development indicators, as thedevelopment expenditure shows a declining trendthroughout the reform period. And the government stillfails to provide safety nets to compensate for thereduction in development expenditure that on the onehand increases unemployment, and the number of thepoor on the other hand. The sequencing of fiscal deficitcuts might have been beneficial if the developmentexpenditure had been further considered in the designingof the programs prior to the reduction in the budget

deficit.

Trade liberalization is supposed to enhance thecountrys income by forcing resources to move from lessproductive uses to more productive uses. But movingresources from low productivity uses to zero productivitydoes not enrich the country, and this is what happened alltoo often under IMF programs. It is easy to destroy jobs,and this is often the immediate impact of tradeliberalization, as inefficient industries close down underpressure from international competition (Stiglitz, 2002).

The immediate reduction in tariff rates and reduction inthe non-tariff barriers before taking adequate measures toenable domestic firms to compete with foreign firms,resulted in the shutting down of domestic units anddeterioration in the current account balance. Anotherpitfall in the sequencing of tariff reforms and changes inthe tax structure is the reliance on the regressive taxstructure. Indirect taxes (Value Added Tax) have increasedthe burden of taxes on the poor and are an importantcontributor to income inequality in the 1990s (Kemal,2003). The major flaw in the sequencing of IMF supported

reforms is that the consideration of fairness is totallyignored. The Washington Consensus policies believe intrickle-down economics, which implies that the best way tohelp the poor is to make the economy grow. What actuallyhappened in Pakistan is that due to inadequateconsideration of poverty in the reforms, poverty increasedvery rapidly in the whole decade of the 1990s causing achronic problem for the Pakistan economy to deal with.

62


29/37


Summary and Conclusion

Pakistan is one of the prolonged users of IMFsupported programs, after initiating it in 1988. Fifteenyears of history with these programs certainly calls for theevaluation of programs. In most parts of the world, theevaluation of the macroeconomic outcomes has been usedas a prime measure to analyze the performance of fund-supported programs. It applies the counterfactualapproach to assess the macroeconomic outcomes of theprogram, which is to compare the macroeconomic

performance of the set of countries with programs to theset of countries without programs or comparing the prereform macroeconomic performance with post reform.

The Generalized Evaluation Estimator technique isused in order to measure the impact of the IMF supportedprograms over the targeted macroeconomic variables. Thetechnique used here takes care of the foreign exogenousfactors, and the other policy options than the IMFsuggested that may affect the performance of targetvariables, by taking the exogenous variables along with

the IMF dummy. The other policy variables are directlyobservable in the absence of the programs but cannot beobserved during the program years, so to overcome thisproblem reaction functions are estimated.

The study suggests that IMF supported programshave worsened the current account balance of Pakistanduring the program years. Immediate liberalization oftrade has caused many domestic units to be shut down.Continuous devaluation of the rupee against the dollar haspushed up the prices of crude oil and machinery, whichare the major inputs of domestic industry and areimported from foreign countries. This increase in the costof the domestic firms has made them unable to competewith other foreign firms. Economic reforms have posed ahuge cost in the form of rapid increases in unemployment.Privatization, reductions in public expenditure combinedwith contractionary monetary policy all have contributedto growing unemployment in the country.

63


30/37


IMF supported programs are also the cause of theincrease in the price level during the last twelve years.Though on the demand side all of the program policieswere mainly contractionary, in order to bring pricestability, the supply side (cost push inflation) has comeinto the picture during program years, which is mainly dueto devaluation and liberalization of trade that hasincreased the cost of the domestic industries by pushingup the prices of inputs used in these industries. Thoughthe effect of fund-supported programs on economicgrowth is negative, it is not statistically significant. But the

IMF programs have succeeded in bringing a slightreduction of the budget deficit during the last twelveyears. Public expenditure cuts and increases in taxrevenue are among the main conditionalities of the IMFeconomic reform packages, but in the case of Pakistan theshare of revenue to GDP has not shown muchimprovement. However development expenditure shows acontinuous declining trend, bringing improvements in thebudget surplus at the cost of increasing unemploymentand rapid increase in poverty.

Countries adoption of gradual stabilization policiesplays a major role in the success of reform programs.Pakistan initiated financial reforms during the early yearsof the programs. The sequencing of financial reforms hasbeen critical in the sense that these reforms wereundertaken before the reduction in the budget deficit.Financial reforms increased the competitiveness of thegovernment in generating funds from the public, resultingin an increase in treasury rates. Increases in the interestrate on the treasury bills and other government securitiescaused the debt servicing of the government to

accelerate. As the government faced the conditionality ofreducing public expenditure an increase in debt servicingput pressure on the government to reduce developmentexpenditure, which resulted in a rapid increase of povertyincidence.

Privatization was adopted, but prior to the adoptionno safety nets were formed for those workers who would

64


31/37


be laid off after the privatization. This wrong sequencing ofprivatization has led to chronic unemployment that is stillincreasing. In the trade regime, immediate openness totrade and reductions in tariffs and other quantitative andnon-quantitative measures in order to enhance efficiencyand competitiveness have caused many domestic units tobe closed, and during these reforms around 3000 unitshave been shut down (Amjad, 2004). The shift in the taxstructure from tariff to regressive taxes has furtherwidened the inequality gap in the country.

References

Ahmad, Mushtaq, 1998, Fiscal adjustment: trade-offs ofmacro-economic goals and recent policy reforms inPakistan. The Pakistan Development Review, 37:4.

Alesina, Albertoand Drazen, Allan, 1991, Why arestabilizations delayed? The American EconomicReview, Vol. 81, No. 5.

Amjad, Rashid, 2004, Solving Pakistans poverty puzzle:who should we believe? What should we do? Paperproceedings of the 19th Annual general meeting andconference, Jan.2004. Pakistan Institute ofDevelopment Economics, Islamabad.

Anderson, M. Torben and Risager, Ole, 1988, Stabilizationpolicys, credibility, and interest rate determination ina small open economy. European Economic Review,33.

Bagci, P., and W. Peraudin, 1997, Do IMF Programs

Work? Global Economic Institutions(Working Paper,1997).

Balassa, Bella, 1982, Structural adjustment policies indeveloping economies. World Development, Vol.10, No. 1.

65


32/37


Ball, Laureence and Sheridan, Niamh, 2003, Doesinflation targeting matters? International MonetaryFund (Working Paper/2003/129).

Baqir, Reza, Ramcharan, Rodney and Sahay, Ratna, 2003,IMF program design and growth: what is the link?International Monetary Fund (Working Paper).

Bengali, Qaiser and Ahmed, Qazi, Mashood, 2001,Stabilization policy vs. Growth-oriented policy:implication for the Pakistan economy. The Pakistan

Development Review, 40:4.

Bhatia, Rattan J., 1975, Fiscal and monetary policy forinternal and external stabilization under fixed andfloating rates in the presence of capitalmovements. The Pakistan Development Review.

Bilquees, Faiz, 1987, The IMF stabilization package andPakistans stabilization experience. The PakistanDevelopment Review, Vol. 26, No. 4.

Branson, H. William and Jayarjah, Carl, 1994, A framework

for evaluating policy adjustment programs: lessonsfrom cross-country evaluation. Proceedings of theWorld Bank Conference on evaluation anddevelopment.

Bruno, Michael, 1992, Stabilization and reforms in EasternEurope: A preliminary evaluation. InternationalMonetary Fund Staff Papers,Vol. 39, No. 4.

Buffie, Edward F., 1992, Stabilization policy and publicsector prices.Journal of Monetary Economics.

Bulir, Ales and Moon, Soojin, 2003, Do IMF-supportedprograms help make fiscal adjustment moredurable? International Monetary Fund (WorkingPaper/2003/38).

Connor, T., 1979, The apparent effects of recent IMFstabilization programs. International Finance

66


33/37


Discussion Paper 135, Board of Governors of FederalReserve System.

Conway, P., 1994, IMF lending programs: Participationand Impact, Journal of Development Economics 45,December 1994.

Cukierman, Alex and Liviatan, Nissan, 1992, Thedynamics of optimal gradual stabilization. TheWorld Bank Economic Review, Vol. 6, No. 3.

De Gregorio J., 1993, Inflation, taxation and long rungrowth.Journal of Monetary Economics, 31(3).

Dornbush, Rudiger, 1991, Credibility and stabilization.Quarterly Journal of Economics, Aug.

Drazen, Allan and Helpman, Elhanan, 1987, Stabilizationwith exchange rate management. The QuarterlyJournal of Economics, November.

Drazen, Allan and Helpman, Elhanan, 1988, The effects ofpolicy anticipations on stablization programs.

European Economic Review, 32.

Franko, H.B., Gustavo, 1990, Fiscal reforms andstabilization: four hyperinflation cases examined.The Economic Journal, 100 (March).

Haq, Ul Nadeem and Khan, Mohsin S., 1998, Do IMF-supported programs work? A survey of the cross-country empirical evidence. International MonetaryFund (Working Paper/98/169).

Helpman, Elhanan and Razin, Assaf, 1987, Exchange ratemanagement: inter temporal tradeoffs. TheAmerican Economic Review, Vol. 77, No. 1.

Hutchison, Michel M., 2001, A cure worse than a disease?Currency crises and the output cost of the IMFadjustment programs. Department of Economics,

67


34/37


Social sciences 1, University of California, SantaCruz.

Iqbal, Zafar, 1994, Macroeconomic effects of adjustmentlending in Pakistan. The Pakistan DevelopmentReview, 33:4.

Ivanova, Anna, Mayer, Wolfgang, Mourmouras, Alex andAnayiots, George, 2003, What determines theimplementation of IMF-supported programs?International Monetary Fund (Working paper

2003/08).

Jahjah, Samir and Montiel, Peter, 2003, Exchange ratepolicy and debt crises in emerging economies.International Monetary Fund (WorkingPaper/2003/60).

Jha, Raghbendra, 2003, Macroeconomics for DevelopingCountries. 2nd edition. Routledge, London and NewYork.

Kemal, A.R., 1994, Structural adjustment, employment,

income distribution and poverty. The PakistanDevelopment Review, 33:4.

Kemal, A.R., 2003, Structural adjustment and poverty inPakistan. MIMAP Technical Paper Series No. 14.Pakistan Institute of Development Economics,Islamabad.

Khan, Mahmood Hasan, 2000, When is economic growthpro-poor? Experiences in Malaysia and Pakistan.International Monetary Fund (Working Paper/ 2002/

85).

Khan, Mohsin S. and Knight, Malcolm, 1985, Fund-supported programs and economic growth.International Monetary Fund, Occasional Paper 41.

Khan, Mohsin S. and Knight, Malcolm. D., 1981,Stabilization programs in developing countries: A

68


35/37


formal framework. International Monetary FundStaff Papers.

Khan, Mohsin S. and Senhadji, S. Abdelhak, 2000,Threshold effect in the relationships betweeninflation rate and growth. International MonetaryFund (Working Paper/2000/110).

Khan, Mohsin S. and Sharma, Sunil, 2001, IMFconditionality and country ownership of programs.International Monetary Fund (Working

Paper/2001/142).

Khan, Mohsin S., 1986, Macroeconomic adjustment indeveloping countries: A Policy Perspective. WorldBank Discussion Paper- Development Policy IssuesSeries.

Khan, Mohsin S., 1990, The macroeconomic effects offund-supported adjustment programs. InternationalMonetary Fund Staff Papers, June.

Khilji, Nasir M. and Leon, Jean Claude, 1989, Output

effects of stabilization policies: the case of Pakistan.The Pakistan Development Review, 28:4.

Kiguel, Miguel A. and Liviatan, Nissan, 1992, The businesscycle associated with exchange rate basedstabilizations. The World Bank Economic Review,Vol. 6, No. 2.

Killick, T., M. Malik and M. Manuel, 1995, What Can WeKnow About the effects of IMF Programmes? WorldEconomy 15, September 1995.

Lizondo, Saul J. and Montiel, Peter J., 1989,Contractionary devaluation in developingcountries. International Monetary Fund StaffPapers.

Loxley, J., 1984, The IMF and poorest countries, Ottawa,Canada: North-South Institute.

69


36/37


Moran, Christian, 1989, Economic stabilization andstructural transformation: lessons from the Chileanexperience, 1973-87. World Development, Vol. 17,No. 4.

Mundell, A. Robert, 1962, The appropriate use of themonetary and fiscal policy for internal and externalstability. IMF Staff Papers.

Naik, Ahmed Ejaz, 1993, Pakistan economic situation andfuture prospects. Pakistan Institute of Development

Economics (Islamabad).

Nsouli, M. Saleh, Rached. Mounir and Funke. Norbert,2002, The speed of adjustment and sequencing ofeconomic reforms: issues and guidelines for policymakers. International Monetary Fund (WorkingPaper/2002/132).

Pakistan, Government of (Various Issues), EconomicSurvey.

Richard, Alan, 1991, The political economy of dilatory

reform: Egypt in 1980s. World Development, Vol.19, No. 12.

Schadler, S. et.al., 1993, Economic adjustment in lowincome countries-experience under the enhancedstructural adjustment facility. InternationalMonetary Fund. Occasional paper 106.

Singer, H.W., 1995, Are the structural adjustmentprograms successful? Pakistan Journal ofDevelopments Economics, Vol. 11.

State Bank of Pakistan, (Various Issues), Annual Report,Karachi

Stiglitz, Joseph, 2000, Capital market liberalization,economic growth and instability. WorldDevelopment.

70


37/37


Stiglitz, Joseph, 2002, Globalization and its discontents,Penguin Books, London.

Uribe, Martin, 1999, Comparing the welfare costs andinitial dynamics of alternative inflation stabilizationpolicies. Journal of Development Economics, Vol.59.

Werner, Alejandro M., 1999, Building consensus forstabilization. Journal of Development Economics,Vol. 59.

Zaidi, S. Akbar, 2000, Issues in Pakistans Economy:Oxford University Press (Karachi).

71

Documents

Nawaz A. Hakro