Abstract

A forecast of the monetary aggregates is an integral part of any financial program. If economic developments relating to national income, the rate of inflation, balance of payments, unemployment, etc., are unfavorable, government policymakers will desire to take corrective action, including changes in monetary management. Usually, before taking any monetary or credit policy measure, the authorities will need some projections or forecasts in respect of the monetary aggregates concerned so that they can compare the likely outcome with and without the policy measure under consideration. When a policy involves limits on the growth of certain monetary aggregates, projections for all the related monetary aggregates would help in determining the overall consistency and the feasibility of the policy.

A forecast of the monetary aggregates is an integral part of any financial program. If economic developments relating to national income, the rate of inflation, balance of payments, unemployment, etc., are unfavorable, government policymakers will desire to take corrective action, including changes in monetary management. Usually, before taking any monetary or credit policy measure, the authorities will need some projections or forecasts in respect of the monetary aggregates concerned so that they can compare the likely outcome with and without the policy measure under consideration. When a policy involves limits on the growth of certain monetary aggregates, projections for all the related monetary aggregates would help in determining the overall consistency and the feasibility of the policy.

Several types of monetary aggregates that appear in the balance sheets of the monetary authorities, the deposit money banks, and the banking system as a whole may be distinguished. Some monetary aggregates, such as net domestic credit of the monetary authorities, are controlled by the monetary authorities themselves. Other monetary aggregates reflect the performances of various sectors of the economy; for example, the change in net foreign assets of the monetary authorities reflects the balance of payments outturn. Still other monetary aggregates usually have close relationships with other variables in the economy; for example, the money stock may be closely related to output and prices.

In order to make the projections realistic, it will be necessary to review the performance of the banking system in the past few years and to observe the outlook or the objectives in relation to the balance of payments and to government budget operations. Developments in output, prices, and possibly other economic aggregates that closely relate to the monetary aggregates would have to be predicted. The projected amounts of monetary aggregates must be consistent with the expected overall economic activity in the future period. Furthermore, the balance sheets that involve the relevant monetary aggregates must be balanced for the forecast period. For example, on the balance sheet of the monetary authorities, the planned expansion of net domestic credit of the monetary authorities plus the desired (or forecast) increase in net foreign assets must be equal to the forecast increase in reserve money. Thus, if this balance sheet identity is not observed, the policy contemplation and the economic objectives are not compatible, given other variables over which the authorities have limited control.

The consolidated monetary account—namely, the monetary survey—is a good basis on which to make monetary projections, as the expected results of government budget operations and of the balance of payments are reflected in this monetary account. Furthermore, monetary policy often involves a restraint on credit expansion. The projected growth of monetary stock (money and quasi-money) consistent with the expected rate of credit expansion can be derived from this account. When the target level for net foreign assets of the banking system, the target rate for growth in income, and other targets are set and the necessary policy measures to achieve these targets are spelled out, the monetary survey in the forecast period becomes a “desired” balance sheet for the banking system and often constitutes a component in a financial program. This will be elaborated in the introduction to Workshop 9 on Financial Programming.

The projections for Kenya to be made in this workshop cover the items in the monetary survey, as well as the demand for monetary stock for December 1977. A simultaneous equation approach to monetary projections is also discussed. This approach implies that both supply and demand factors are at work in determining the levels of monetary stock.

RECENT MONETARY DEVELOPMENTS IN KENYA

After six years of steady growth in international reserves, Kenya experienced its first balance of payments deficit in 1971. The authorities responded by imposing selective credit controls in July 1971 and by instituting an extensive import control system in January 1972. The selective credit controls included absolute ceilings on credit to importers of consumer goods; these were set at 70 per cent of the amounts then outstanding for motor vehicle importers and 95 per cent for importers of other consumer goods. In February 1972, a ceiling of 12 per cent per annum on the growth of commercial bank credit to the private sector was established.

During the 18 months ended June 1973, imports declined and net foreign assets rose by over 60 per cent, while credit to the private sector increased by only 6 per cent. However, credit to the Cereals and Sugar Finance Corporation (CSFC) and to the Agriculture Finance Corporation (AFC) and net credit to the Government of Kenya rose by 64 per cent. Money and quasi-money grew at an annual rate of 20 per cent for this period.

As a result of these improvements in the balance of payments, the ceiling on bank credit expansion was lifted in March 1973, most of the selective credit controls were abolished in April, and import controls were liberalized in July. During the 12 months to July 1974, private sector credit rose by almost 50 per cent, net foreign assets dropped by 20 per cent, money supply increased by 19 percent, and despite a rapid increase in bank deposits, bank reserves fell by more than 50 per cent. By mid-1974, commercial banks were approaching the minimum liquidity ratio of 15 per cent.

On June 1, 1974, the Central Bank of Kenya forecast a large loss in foreign exchange reserves over the coming 18 months because of the anticipated sharp increase in import prices. Consequently, the following measures were taken. First, a 12 per cent ceiling on the increase in bank credit to the private sector and public enterprises (excluding the CSFC and the AFC) was imposed for the period June 1974 to June 1975. Second, the 15 per cent liquidity ratio was extended to licensed financial institutions, except for housing finance companies. Third, local borrowing for nonresident-controlled companies was further restricted. On July 1, 1974, the Central Bank raised the legal minimum interest rate on time and savings deposits from 3 to 5 per cent and the minimum lending rate from 7 to 8 per cent. The discount rate on crop finance paper was raised from 6.5 per cent to 7 per cent.

From June 1974 to June 1975, money plus quasi-money rose by 6 per cent, owing to an increase of 21 per cent (KSh 908 million) in domestic credit, as net foreign assets fell by KSh 580 million or 40 per cent. Credit to the nongovernment sector increased by only 7 per cent (KSh 274 million) during this period as a result of the more restrictive credit policy. However, net credit to the Government increased sharply by KSh 634 million (Table 1). The composition of money plus quasi-money shifted as the share of quasi-money in the total rose from 28 per cent in June 1974 to 33 per cent in June 1975, mainly because of the increase in the legal minimum interest rate on time and savings deposits.

Table 1.

Kenya: Monetary Survey, 1973–77

(In millions of Kenya shillings)

article image
Sources: Central Bank of Kenya, Annual Report for the Financial Year Ended 30th June, 1977; and International Monetary Fund, International Financial Statistics, July 1978.

Foreign liabilities include use of Fund credit.

Foreign exchange holdings.

Includes Cereals and Sugar Finance Corporation and Agricultural Finance Corporation.

Adjusted for “uncleared effects.”

The domestic counterpart of Kenya’s SDR allocation was transferred in September 1975 to treasury deposits with the Central Bank of Kenya.

In mid-1975, in view of the slackened pace of economic activity and a more favorable foreign reserve position than anticipated, the Central Bank sought to encourage a more rapid credit expansion than had been occurring up to that time. The Central Bank expressed willingness to lend to the commercial banks. To promote lending to the agricultural sector, the Central Bank required the commercial banks to increase the share of their deposits lent to the agricultural sector to 17 per cent by end-June 1976, compared with 10 per cent at end-June 1975.

During the second half of 1975, money and quasi-money grew at an annual rate of 24 per cent, owing to the increase in domestic credit, as net foreign assets declined slightly. Net credit to the Government increased sharply by KSh 667 million. The domestic counterpart of Kenya’s allocation of special drawing rights (SDRs) by the International Monetary Fund (KSh 134 million) was transferred in September 1975 to treasury deposits with the Central Bank. Credit to the nongovernment sector increased by KSh 450 million, equivalent to an annual rate of 21 per cent during these six months.

In 1976, the balance of payments registered a surplus of KSh 776 million, compared with a deficit of KSh 356 million for the previous year. Export receipts increased by 31 per cent. In anticipation of a rapid growth in bank deposits due to the increase in export receipts, the Central Bank increased the liquidity ratio of commercial banks from 15 per cent to 18 per cent, effective July 1, 1976. Nevertheless, money and quasi-money increased by 24 per cent in 1976. In October, in view of the low rate of credit expansion to the private sector during the first three quarters, the Central Bank informally encouraged the commercial banks to increase their lending to the private sector. Furthermore, effective December 8, 1976, the Central Bank abolished the minimum lending rate of 8 per cent. Accompanying these measures, the Central Bank reduced its discount rates by a half of 1 per cent. For 1976 as a whole, domestic credit increased by 16 per cent (KSh 1,061 million), reflecting partly the response to monetary policies and partly the general economic recovery toward the end of the year.

PROJECTIONS

For 1977, the balance of payments forecasts implied an overall surplus of KSh 2,200 million.1 Central bank domestic credit for 1977 was to be reduced by KSh 800 million. The overall budget deficit for fiscal year 1978 was estimated at KSh 2,384 million, compared with a deficit of KSh 1,327 million for fiscal year 1977. External borrowing was to finance 29 per cent of the estimated deficit, compared with 35 per cent in the previous fiscal year. Net borrowing from the banking system was expected to be KSh 1,200 million, compared with minus KSh 5 million for the previous fiscal year. The remainder of the budget deficit was to be financed through the nonbank domestic sector.

Supply of Money and Quasi-Money

For the period 1967–76, regression equations for the supply of money and the supply of money plus quasi-money have been estimated as follows:2

MOt=910.55(1.74)+3.0870RMt(8.04)R¯2=0.876(1)
MQt=1479.23(1.84)+4.6052RMt(7.79)R¯2=0.869(2)

where

MO = money at the end of December, in millions of Kenya shillings

MQ = money plus quasi-money at the end of December, in millions of Kenya shillings

RM = reserve money at the end of December, in millions of Kenya shillings.

The figures in parentheses indicate the t values for the corresponding coefficients.

Demand for Money and Quasi-Money

Demand functions for money and for money plus quasi-money in real terms for Kenya for the period 1967–76 have been estimated as follows:

MOt*=1883.91(3.05)+0.3706(6.42)GDPt28.9752(1.58)P*tR¯2=0.929(3)
MOt*=2630.84(3.88)+0.5243(8.28)GDPt34.9792(1.74)P*tR¯2=0.959(4)

where

MO* = two-year averages of end-of-year stocks of money, in millions of Kenya shillings, deflated by the gross domestic product (GDP) deflator

MQ* = two-year averages of end-of-year stocks of money plus quasi-money, in millions of Kenya shillings, deflated by the GDP deflator

P = percentage change in the GDP deflator

GDP = GDP at factor cost at constant 1972 prices, in millions of Kenya shillings.

The growth rate in real GDP has been targeted at 7 per cent for 1977, and the corresponding growth rate for nominal GDP has been projected to be 25 per cent.

Projections with a Simultaneous Equation Model

As mentioned earlier in this chapter, the monetary stock can be estimated by using a simultaneous equation model that includes both supply and demand factors. The simplest form of such a model is:

MOtd=kMGDPt+ut(5)
MOts=mRMt+vt(6)
RMt=NFAt*+NDCt*(7)

where

MOd = demand for nominal stock of money

MOs = supply of nominal stock of money

MGDP = GDP at current prices

NDC* = domestic credit of the monetary authorities

NFA* = net foreign assets of the monetary authorities.

With MGDP and NDC* exogenous and MOd = MOs = MO in equilibrium, the above model determines MO, RM, and NFA*.

A less aggregative model, which involves estimations of imports, domestic aggregate expenditure, nominal GDP, change in net foreign assets of the banking system, and monetary stock, is provided as follows:

Mtd=a0+a1AEt+ut(8)
ΔMt=α[MtdMt1](9)
AEtd=b0+b1MQt+b2MGDPt+vt(10)
ΔAEt=β[AEtdAEt1](11)
MGDPt=AEt+XtMt(12)
ΔNFAt=XtMt+CMt(13)
ΔMQt=ΔNFAt+ΔNDCt(14)

This model assumes (1) that import demand (Md) depends on aggregate expenditure (AE); (2) that change in actual imports (ΔM) adjusts to excess demand for imports; (3) that desired aggregate expenditure (AEd) depends on nominal GDP (MGDP) and monetary stock (MQ); (4) that change in actual value of aggregate expenditure adjusts to the difference between desired expenditure and actual expenditure; and (5) that change in monetary stock reflects the balance of payments outturn (ΔNFA) and change in domestic credit of the banking system (ΔNDC). Equations (12) through (14) are accounting identities. Exports (X), domestic credit (NDC), and net foreign capital inflows (CM) are treated as exogenous. The structure of the model emphasizes the interrelationships between monetary variables, aggregate expenditure, and the balance of payments.

Using data for Kenya for the period 1967–76, a reduced-form equation for monetary stock derived from the above model is as follows:

MQt=650.5-0.1477Mt-1-0.2745AEt-1(15)+0.8776NFAt-1+0.8122Xt+0.8776NDCt+0.8776CMt-2165.36Dt

In estimating this model, a dummy variable (D) has been included in the equations for imports and aggregate expenditure to reflect the exceptionally large increases in imports and inventory accumulations in 1974 due to the speculation of a further sharp increase in import prices.

EXERCISES

Exercise A

  • On the basis of the above information, projections are to be made for domestic credit, net foreign assets, money, and quasi-money for the banking system for December 1977 (Table 1). The following steps may be followed:

  • 1. Net foreign assets. Information can be obtained from the statement of the balance of payments forecasts for 1977 above. Net foreign assets at the end of 1977 should be equivalent to net foreign assets at the end of 1976, plus the change in net foreign assets reflecting the balance of payments outcome in 1977. Please note that net foreign assets of the commercial banks are not included in official foreign reserves of Kenya.

  • 2. Money and quasi-money. Projections for money and quasi-money are to be made using regression equations (1) and (2). The following balance sheet identity is required for the projections:3
    ΔRMt=ΔNFAt*+ΔNDCt*
    The change in net foreign assets for the banking system has been projected in item 1 above. The division between net foreign assets of the monetary authorities and those of the commercial banks may be made on the basis of the situation at the end of December 1976 (Table 1). The level of domestic credit of the monetary authorities at the end of December 1976 can be obtained from Table 2.
  • 3. Other items net. This item may be assumed to stay the same as in December 1976. Alternatively, a trend ratio between this item and the total liabilities of the banking system may be assumed.

  • 4. Domestic credit. This item may be taken to be the difference between money plus quasi-money plus other items (net) and net foreign assets of the banking system. Information about net claims on the Government is contained in a statement concerning the fiscal year 1978 budget in the preceding section. For this exercise, net borrowing from the banking system by the Government for calendar year 1977 may be taken as the average of net borrowing for fiscal years 1977 and 1978. Credit to the nongovernment sector is to be taken as the difference between total domestic credit and net domestic credit to the Government.

Table 2.

Kenya: Summary Accounts of the Monetary Authorities, 1973–76

(in millions of Kenya shillings)

article image
Sources: Central Bank of Kenya, Annual Report for the Financial Year Ended 30th June, 1977; and International Monetary Fund, International Financial Statistics, July 1978.

Includes balances with banks abroad and foreign securities.

Includes use of Fund credit.

Includes revaluation account, uncleared effects, and other unclassified assets and liabilities.

The domestic counterpart of Kenya’s SDR allocation was transferred in September 1975 to treasury deposits with the Central Bank of Kenya.

Exercise B

Using regression equations (3) and (4), make a projection for the demand for money and quasi-money for the same date as in Exercise A, based on the targeted income and prices. Historical data for gross domestic product and prices are available in Table 3.

Table 3.

Kenya: Selected Statistical Data, 1966–76

article image
Source:

ISSUES FOR DISCUSSION

  • 1. What are the meanings of the coefficients to reserve money (RM) in regression equations (1) and (2)P In relating money or money plus quasi-money to reserve money as in the above equations, the reserve ratios, the ratio between currency and demand deposits, the ratio between time deposits and demand deposits, and the ratio between excess reserves and demand deposits are treated as parameters. How would changes in these ratios affect the coefficients to reserve money in the equations?

  • 2. An alternative presentation of the relationship between money and reserve money is to treat the money multiplier as a composite variable. From the following two identities:

MO=CY+DDRM=CY+Rd+Rt+Re

it follows that

MO=CY+DDCY+Rd+Rt+ReRM

where

CY = currency outside banks

DD = demand deposits

Rd = reserves against demand deposits

Rt = reserves against time and savings deposits

Re = excess reserves.

The ratio in the above expression is equivalent to the money multiplier.

Compare this approach with the approach implied in equations (1) and (2).

  • 3. The Kenyan authorities have responded to the balance of payments deficit by imposing credit and import restrictions (1971, 1972, 1974y and 1975). Discuss and evaluate, on the basis of historical data, (a) the effectiveness of the measures in improving the balance of payments and (b) the impact of these measures on real gross domestic product (GDP).

  • 4. Comment on the method used to forecast the demand for money. Have important explanatory variables been neglected?

  • 5. Compare the projection for money and quasi-money obtained in Exercise B with that in Exercise A and explain the implications of the discrepancy.

  • 6. The actual and fitted values for the real money plus quasi-money, based on regression equation (4) are as follows:

    article image
    Was there significant “excess liquidity” or “deficient liquidity” in the economy in any period? Would there be any excess liquidity if no corrective measure was taken in 1977? Explain the implications of the existence of excess liquidity with reference to the case of Kenya (for example, the impact on the balance of payments).

  • 7. In the model presented in equations (5) through (7), money stock, reserve money, and net foreign assets of the monetary authorities are all determined by the same factors—namely, MGDP, which is the determinant of the demand for money, and NDC*, which is a policy variable. This approach to net foreign assets is sometimes called the monetary approach. Compare this approach with the approach implied in the model presented in equations (8) through (14).

  • 8. Is it possible in the reduced-form equation (15) to identify “demand factor” variables and “supply factor” variables? Explain.

  • 9. Discuss the merits and demerits of a forecast based on equation (15) versus forecasts based on Exercises A and B.

1

In January 1977, the official projection for the balance of payments was a surplus of KSh 445 million. The official projection has been revised for the purpose of this workshop.

2

Selected data for Kenya for the period 1966–76 appear in Table 3.

3

Assuming that other items (net) in the summary accounts of the monetary authorities remain the same.