Chapter

V Credit Policy and Developments, 1974–82

Author(s):
Rattan Bhatia
Published Date:
May 1985
Share
  • ShareShare
Show Summary Details

As noted in the Section IV, until recently the BCEAO continued to rely mainly on its rediscount policy to influence monetary developments and, to some extent, on the mechanism of prior authorization to determine the sectoral allocation of credit. Nevertheless, under the new statutes and the rules of intervention, the BCEAO has acquired a wide range of instruments, so that it now can adopt discretionary policy for individual member countries.

The new decentralization of monetary policy has allowed greater flexibility to domestic monetary authorities. Within overall monetary policy, interest rate structures, and exchange rate policy, national authorities have been able to use different instruments appropriate to their particular situation.

Thus, although the BCEAO has not imposed overall reserve requirements on banks, a voluntary reserve deposit arrangement was worked out in Ivory Coast during 1977–79. According to this agreement, the Price Stabilization Fund—the banking department of the Caisse Autonome d’Amortissement (CAA)—agreed to deposit a major part of its increasing liquidity with the CAA, which in turn agreed to maintain these resources in a specially remunerated time deposit account with the BCEAO. The agreement was aimed at sterilizing part of the rapidly rising liquidity of the banking system emanating from the boom in exports.

In Togo, sectoral guidelines on credit have been introduced and are announced annually. Niger uses the system of prior authorization, but without a firm set of sectoral priorities. A system of monthly credit ceilings operates in Burkina Faso and Senegal. In Senegal, the regulatory and allocative functions of the Central Bank were initially not effective: the rediscounting policy continued to be as accommodating as under the old statutes, and the Central Bank was reluctant to apply credit controls or penalties on banks exceeding their limits on credit expansion. The share of bank loans in total bank credit subject to prior authorization in Senegal also declined markedly. However, in 1979 the Central Bank imposed monthly credit ceilings on individual banks—a system which was also introduced in Burkina Faso in January 1980—and these have been more effective.

In 1975 the BCEAO initiated the setting of comprehensive annual targets for monetary aggregates. In October each year the National Credit Committee of each member country is required to prepare its proposals for monetary aggregate targets for the coming year. In general, each Committee projects the rate of growth, the rate of inflation, the evolution of the external sector position, and the budgetary financing requirements for the following year. In light of the projections for growth and inflation and objectives for external reserves and budgetary requirements, the rates of growth of the monetary aggregates are determined. These targets are then submitted to the BCEAO headquarters in Dakar, where they are considered and, if necessary, modified to conform to the overall monetary policy of the Union.

The first set of targets was prepared for 1976 (see Table 13 and Statistical Appendix Tables 14 to 19). Between 1976 and 1978, the targets for domestic liquidity for the region as a whole were exceeded, while in 1979–80 the rate of growth of domestic liquidity remained well below the targets. The main factor accounting for the excess growth of domestic liquidity in the earlier period was a sharp expansion in domestic-credit, particularly to the private sector. By comparison, in 1977 and 1978 net credit to governments was considerably below targets. During 1976 and 1977 the expansionary impact of domestic credit also tended to be reinforced by net foreign assets, which increased beyond target levels, but after 1978 net foreign assets were lower than projected. Actual credit to the private sector exceeded the targets in the later period, as they had earlier, but in 1979 and 1980 the excess was more than fully offset by higher net government deposits relative to targets. Net foreign outflows rose to CFAF 140 billion and CFAF 362 billion in late 1979 and 1980 respectively, compared with targets of CFAF 21 billion and CFAF 102 billion. It was this decline in net foreign assets, together with the slowdown in domestic credit to the public sector, which brought the rate of growth of domestic liquidity down below target levels.

Data on monetary developments in individual countries also reveal similar general patterns. In most countries, domestic liquidity exceeded target levels through 1978, but generally remained below target in 1979 and 1980. (The two notable exceptions were Benin and Burkina Faso: domestic liquidity in Benin exceeded target levels only in 1976 and 1980; while in Burkina Faso targets were exceeded only in 1976.) Moreover, in almost all countries, the target for credit to the private sector was exceeded in most years, although the excess differed considerably from country to country and from year to year. Net credit to governments was also generally below target levels. Finally, the results on the achievement of targets for net foreign assets varied considerably. In Burkina Faso. Ivory Coast, and Senegal, net foreign assets declined more than projected, while in Benin and Togo they were in most years above target. In Niger, net foreign assets exceeded targets during 1976–78 but fell below them in 1979 and 1980.

Table 13.BCEAO: Monetary Survey—Targets and Actuals, 1975–80(In millions of CFA francs)
197519761977197819791980
TargetsActualsTargetsActualsTargetsActualsTargetsActualsTargetsActualsTargetsActuals
Net foreign assets-4.3004.81415.10022.96433.3001.47421.400-139.650-101.800-362.065
Net domestic credit459.500529.600636.243678.900824.335849.900989.8811130.7001148.1471438.3001407.095
Net credit to government-41.400-21.400-3.996-23.700-87.851-45.800-111.514-47.500-129.909-35.500-49.260
Credit to the private sector500.900551.000640.239702.600912.186895.7001101.3951178.2001278.0561437.8001456.355
Other items (net)-22.900-30.500-44.886-18.700-35.852-25.100-50.815-71.600-64.583-84.800-20.248
Domestic liquidity432.300499.100596.171675.300811.560858.100940.5401080.500943.9141251.7001024.782
Note: This table is derived as a summation of the individual country’s balance sheet of the banking system for the purpose of consistencySources: Information provided by the BCEAO: and International Monetary Fund. International Financial Statistics (Washington: IMF).
Note: This table is derived as a summation of the individual country’s balance sheet of the banking system for the purpose of consistencySources: Information provided by the BCEAO: and International Monetary Fund. International Financial Statistics (Washington: IMF).

Monetary developments in the Union are shown in Tables 14 and 15. The ratios of conventionally denned “narrow” money (M1 and “broad” money (M2) to GDP have fluctuated over time. (M1 is defined as currency plus demand deposits; M2 as M1 plus time deposits.) Following a period of particularly pronounced volatility in the mid-1970s, the ratio of M1 to GDP stabilized somewhat, while that of M2 to GDP continued to fluctuate with an increasing trend rate. The ratio of M1 lo GDP was higher in 1981 than it was in 1963 in all countries, and was lower than its 1974 level only in Ivory Coast and Niger. The difference between the smallest ratio and the largest ratio was larger in 1981 than in 1963 or 1974.

Table 14.WAMU and Members: Narrow Money as a Percentage of GDP, 1963–83
YearAverageBeninBurkina FasoIvory CoastNigerSenegalTogo
196314.8814.168.8417.238.8517.6311.48
196415.1413.868.7216.938.6419.0412.14
196514.3012.998.7617.629.0715.0612.00
196613.4713.248.8617.997.2413.1210.81
196713.3012.428.6817.607.5412.6011.70
196814.0713.529.4318.107.5813.1012.60
196914.9014.8610.0219.088.7312.8713.92
197015.8615.689.5020.128.6814.7413.83
197116.4517.279.4220.929.6914.5614.87
197216.6719.149.6421.8310.4713.4113.75
197318.0317.9912.2720.8612.3819.1711.80
197421.0221.9714.2321.7116.4824.2319.30
197519.6623.8217.7321.5512.9018.5017.49
197621.0019.0819.9423.4110.2720.6624.60
197722.3319.3918.3124.8911.2322.5624.19
197822.1918.4018.4523.3112.8125.5828.56
197920.1817.4513.3222.3012.9420.8624.83
198019.0120.1914.6319.7512.2221.9823.45
198120.0320.0816.2219.9912.5524.3730.58
198219.1423.9315.5918.4610.9222.3933.36
198318.1021.5915.0918.249.8620.0529.53
Note: M1 is defined as currency plus deposits.Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Note: M1 is defined as currency plus deposits.Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Table 15.WAMU and Members: Broad Money as a Percentage of GDP, 1963–83
YearAverageBeninBurkina FasoIvory CoastNigerSenegalTogo
196315.5214.509.0718.368.9118.2211.78
196416.8314.369.0220.658.6419.5912.45
196515.5113.238.9520.389.0715.6012.53
196614.6913.408.9820.917.6413.4611.09
196714.8412.988.7621.128.0912.9712.23
196816.1214.229.6122.228.3513.6914.30
196917.9315.5210.2625.149.4213.6716.51
197018.9116.349.7325.719.4415.9317.28
197119.6618.379.8826.7211.1215.7417.66
197219.2320.9310.3226.0011.8614.6816.16
197322.0721.5913.1726.1614.0922.7116.24
197426.6525.7715.2329.7718.9927.6223.44
197524.8528.2519.1729.3114.2921.1922.93
197626.8722.8122.8331.4812.1224.7430.78
197729.1923.0921.1634.0713.0727.0831.90
197829.7523.1522.8132.6214.9732.1138.57
197926.0121.0717.7229.1114.5727.7331.37
198024.8527.3018.6826.1914.7528.3230.75
198126.7425.0621.1727.5315.7932.3938.46
198226.0928.2320.4726.4912.7731.0843.40
198325.0725.5420.0225.8712.2628.9341.86
Note: Broad money is defined as narrow money, plus time deposits.Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Note: Broad money is defined as narrow money, plus time deposits.Source: International Monetary Fund, International Financial Statistics (Washington: IMF).

The ratio of M2 to GDP has shown larger fluctuations and a greater tendency to rise, mainly because of a substantial increase in the “savings habit” that occurred in the WAMU countries throughout the 1970s. Time and savings deposits in all the WAMU countries show an unsteady but continuing increase (Statistical Appendix, Table 20). Although all six countries began with a roughly similar level of time deposits in 1963, the rate of expansion has differed considerably among countries, as reflected in the volatility of the ratio of M; to GDP. The main reason for this was a sharp increase in savings and the monetization of the economy, which had more influence on the BCEAOs monetary liabilities than any variations in the money multiplier, as observed in Section IV. The habit of banking is an institutional and sociological phenomenon and varies considerably among countries, often causing divergences from monetary targets despite the BCE AO’s substantial control over credit and the money supply in the Union. This is evident from the fact that the 1974 changes have had no impact on the changes in the ratio of M2 to GDP during the post-reform period. Ivory Coast, the important beneficiary of time deposit growth, and Benin and Burkina Faso (the laggards in this regard) have had considerably different experiences with their M2 to GDP ratios even after the 1974 changes.

Since 1974, the BCEAO has supplemented the re-discounting mechanism by money market operations—credit allocations under which it exercises its power of prior approval of bank loans exceeding a certain amount. Prior approval is required for loans larger than CFAF 100 million in Ivory Coast, CFAF 70 million in Senegal, and CFAF 30 million in the other member countries. In addition, starting with Togo in 1976, commercial banks are required to allocate a certain minimum proportion of credit to priority activities. In Senegal in 1979 the BCEAO introduced a monthly credit ceiling on individual banks to control the expansion of credit: the practice has since been introduced in all member countries. The “voluntary” arrangement mentioned earlier that was worked out in 1977 in Ivory Coast between the BCEAO and the Price Stabilization Fund of the CAA lapsed in the third quarter of 1979 when these deposits were completely withdrawn.

The impact of rediscounts and of money market operations on credit developments during this period has been unpredictable. Tables 16 and 17, which give ratios of changes in private sector credit to changes in rediscounts and to changes in rediscounts plus net money market positions since 1975, show that neither of these two ratios has been stable, nor has either been consistently positive or negative. Relatively stable relations would suggest that these two instruments could be effectively relied upon to control private sector credit, but they clearly cannot be considered sufficiently consistent or predictable.

Table 16.WAMU Members: Ratio of Changes in Private Sector Credit to Changes in Rediscounts, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19752.946.30-14.113.8818.444.25
19760.492.88-3.520.1126.106.69
19776.079.153.94-18.377.719.17
19787.821.696.517.462.1140.31
19793.03-0.441.742.402.331.06
19802.762.580.95-3.351.545.62
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Table 17.WAMU Members: Ratio of Changes in Private Sector Credit to Changes in Rediscounts and Net Money Market Positions, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19752.8830.50-19.213.8818.324.25
19760.492.406.130.1114.77-5.75
19772.968.833.53-2.389.114.45
19783.991.5123.6113.132.06-2.41
19794.49-0.404.791.842.480.90
19802.763.510.92-20.051.542.04
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).

The relative importance of money market operations compared with the changes in total net financing (rediscounts plus net money market advances) has also varied considerably among countries (Table 18). In Benin, Ivory Coast. Niger, and Togo money market operations seem to have played an important role recently in supplying funds lo or rechanneling them from commercial banks, although the size of their contribution has varied considerably. In Benin about 50 percent of total net financing was provided by the money market in 1979, while in 1980 its role was negligible. In Ivory Coast in 1976, the change in net money market borrowing was nearly three times larger than the change in net financing, implying a considerable drop in rediscounts. In 1977, much greater reliance was placed on rediscounts, bringing the ratio down to 10 percent. The situation was reversed in 1978, with net deposits in commercial banks rising sharply, to 2.6 times total net financing. The same pattern repeated itself in 1979, although it was less pronounced. In 1980 net money market operations fell to a small proportion of net financing. In Niger in 1977, they accounted for most of the financing, while in 1978 and 1980 (particularly the latter) there was a sharp rise in net deposits from the BCFAO. In Togo, net money market operations accounted for a substantial part of financing in every year except 1979.

Table 18.WAMU Members: Ratio of Changes in Net Money Market Positions to Changes in Total Net Financing, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19750.01-3.83-0.36
19760.010.162.740.431.86
19770.050.030.100.87-0.180.51
19780.490.10-2.62-0.760.021.06
1979-0.480.08-1.750.23-0.060.15
19800.360.03-4.970.63
Note: Total net financing consists of rediscounts plus net money market operations. A negative sign indicates net deposits in commercial banks.Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Note: Total net financing consists of rediscounts plus net money market operations. A negative sign indicates net deposits in commercial banks.Source: International Monetary Fund. International Financial Statistics (Washington: IMF).

In the other countries, the importance of the money-market in refinancing remained relatively small. In Senegal it ranged between 2 percent and 18 percent except in 1976. In Burkina Faso it was relatively unimportant, although in 1975 there was a large net buildup of deposits, it is clear that the importance of the money market in channeling funds has differed among countries, and has been more crucial in Benin. Ivory Coast, Niger, and Togo, than in Burkina Faso and Senegal.

The explanation for the limited efficacy of the monetary tools used by the BCEAO—the rediscount mechanism and money market operations—in realizing monetary targets, whether for credit or for external reserves, is probably the same as the reason the rediscount mechanism was ineffective in the pre-1975 period. As long as the BCEAO relies on its refinancing facilities, its influence is circumscribed by the ability of the commercial banks to finance their credit expansion and by the public’s propensity to hold banks’ deposit liabilities. In addition, although the commercial banks’ recourse to rediscounts is no longer automatic under the revised intervention rules, the introduction of preferential interest rates for priority sectors obliges the BCEAO to continue to offer rediscount facilities for such credit.

The share of domestic private sector credit financed by the BCEAO—-which includes preferential credit to priority sectors—averaged about 17 percent during the four-year period 1977–80 (Table 19). The BCEAO refinanced practically all “seasonal” credit. Not all of this, however, is seasonal in its traditional sense. Some governments have increasingly passed on the financial burden of some of their social policies (such as consumer subsidies) to the national price stabilization funds, which engage mainly in marketing export crops. These stabilization funds have, therefore, often covered their losses from these subsidies by “seasonal” borrowing from banks, which they accumulate from one year to another. (The most striking example of this was borrowing by the ONCAD (abolished in 1980) in Senegal.) This blurring of the distinction between seasonal and nonseasonal credit led the BCEAO to remark in one of its publications that the recent credit developments “confirm the rigidity at the base of certain categories of financing which no longer follow the flows traditionally observed in this period between seasons.”10

Table 19.BCEAO: Refinancing of Bank Credit, 1977–80(In billions of CFA francs, end of September)
1977197819791980
Bank credit to private sector8811,0681,2471.405
Seasonal126120142183
Ordinary7559481,1041,222
BCEAO refinancing234281327416
Seasonal115111011351178
Ordinary119171192238
Percentage of ordinary credit refinanced16181719
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).

Estimates.

Source: International Monetary Fund. International Financial Statistics (Washington: IMF).

Estimates.

The fact that the BCEAO’s share in the financing of ordinary credit has remained below 20 percent indicates the relative impotence of the rediscounting mechanism in controlling monetary developments. This remains true even when account is taken of the BCEAO’s discretion to decide whether a commercial bank’s demand for funds should be met through the rediscount window or from the money market. Until mid-1979, interest rates in the money market were lower than the BCEAO’s basic rediscount rate, but higher than its preferential rediscount rate. Between August 1979 and April 1980, when the structure of interest rates was revised upward, the money market rates equalled or exceeded the basic rediscount rate, but the April revision restored the earlier discrepancy between the BCEAO’s basic rediscount rate and the money market rates.

Although the interest rate structure relating to the BCEAO’s refinancing remained unchanged between July 1975 and April 1980, the Bank did allow some variation in the money market rates. However, the demand for funds in the money market tended to exceed the supply and the Bank’s influence in restraining credit is not certain. This is especially true since both the creditor and debtor interest rates applied by the deposit money banks (also fixed by the BCEAO) remained unchanged until the 1980 revision.

Another limitation on the BCEAO’s ability to control domestic liquidity via rediscount ceilings (and, therefore, reserve money) is the unstable behavior of the domestic money multiplier (the ratio of money and quasi-money to reserves). During 1975–80 this multiplier ranged from 2.5 to 3.3 for Senegal, 2.5 to 3.1 for Ivory Coast, 2.1 to 2.8 for Burkina Faso, and 2.2 to 2.6 for Togo (Table 20). Only in the case of Niger was the multiplier relatively stable. In the circumstances, even if the authorities had been able to control the domestic part of reserve money through the rediscount policy, there could have been a variation in the target of domestic liquidity of up to 33 percent due to the autonomous movements in the domestic liquidity multiplier.

Table 20.WAMU Members: Ratio of Domestic Liquidity to Reserve Money, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19753.652.072.472.212.492.40
19762.982.362.912.062.932.17
19773.332.263.142.182.932.41
19783.902.792.642.042.842.51
19792.832.322.972.153.302.64
19803.422.462.992.242.992.52
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).

The effectiveness of the rediscount mechanism has been further restricted by the freedom allowed to commercial banks to use other sources of financing, in particular foreign financing, to supplement domestic deposits, rediscounts, and net money market operations to meet the demand for credit. Although in some countries in some years the data show a reduction in net or gross foreign liabilities, in general these ratios have been positive (Tables 21 and 22), In some countries, notably Niger, the monetary authorities have attempted to contain foreign borrowing by commercial banks through moral suasion and by limiting rediscounts to banks borrowing abroad for the purpose of credit expansion.

Table 21.WAMU Members: Ratio of Changes in Private Sector Credit to Changes in Net Foreign Liabilities, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
1975-5.47-10.251.4010.423.070.63
19760.194.75-79.80-2.72-3.47-5.17
1977-2.862.85-15.65-4.777.393.87
1978-5.44-12.322.604.146.76-3.01
1979184.131.274.5128.102.95-0.90
198023.87-17.272.211.36-14.140.77
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Source: International Monetary Fund. International Financial Statistics (Washington: IMF).
Table 22.WAMU Members: Ratio of Changes in Private Sector Credit to Changes in Gross Foreign Liabilities, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19752.06-11.908.6570.992.653.67
19760.035.775.01-0.37-91.429.35
197730.052.52-33.832.144.822.83
19782.25-11.933.616.507.911.24
19797.804.006.594.802.63-1.00
1980-30.582.022.072.03-103.041.19
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).

As mentioned earlier, the BCEAO has also used the mechanism of prior authorization, especially in Senegal and Togo. Although the original intention of this instrument was to influence the sectoral allocation of credit, in Senegal the authorities have also used it to try to influence the quantity of credit. However, its effectiveness is not apparent. For example, in Togo, where sectoral guidelines for credit expansion were introduced in 1976 and have been announced each year, the three priority sectors still received less than half the share of total credit that the guidelines suggested. In Senegal, the share of bank loans with prior authorization has declined from slightly over two thirds in 1975 to slightly over half of the total credit extended by the deposit money banks. The BCEAO has been reluctant to apply the penalties to banks not meeting the BCEAO’s guidelines.

Data on reserves show that, from 1975 to 1980, banks in most countries appear to have maintained as low a level of reserves as possible (Table 23). In most countries the ratios of reserves to demand and time deposits have been maintained at about 5 to 10 percent: only in Togo has the ratio been consistently and, on occasion, considerably above 9 percent. The expansion in credit to the private sector, however, has generally exceeded the expansion in demand and time deposits (see Statistical Appendix, Tables 21 and 22). The ratio of changes in credit to the private sector relative to changes in demand and lime deposits has also fluctuated considerably (Table 24). The instability of these ratios, in addition to the practice of foreign funding by commercial banks, hampers the BCEAO’s task of controlling monetary developments through rediscount ceilings.

Table 23.WAMU Members: Ratio of Reserves to Demand and Time Deposits, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19750.010.080.050.050.080.09
19760.020.030.040.040.060.16
19770.020.040.070.100.060.12
19780.120.050.130.200.080.11
19790.050.070.060.060.050.10
19800.030.050.064.080.070.08
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Table 24.WAMU Members: Ratio of Changes in Credit to the Private Sector to Changes in Demand and Time Deposits, 1975–80
YearBeninBurkina FasoIvory CoastNigerSenegalTogo
19751.471.495.153.742.34-2.13
19760.132.051.03-0.200.630.85
19771.576.981.410.561.992.25
19781.381.183.071.792.510.51
1979-5.212.721.138.391.740.91
19801.260.5110.031.435.072.35
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).
Source: International Monetary Fund, International Financial Statistics (Washington: IMF).

A final instrument of monetary control relates to bank credit to government, which, as mentioned earlier, is statutorily limited to 20 percent of the previous year’s tax receipts. This limit applies to gross credit, however, and the BCEAO has no control over governments’ deposits with the banking system. Also, although in theory the Council of Ministers approves overall country credit targets (including credit to government), the recourse of governments to bank credit is automatic, subject to the statutory ceiling. These factors have tended to increase the amplitude of net bank credit ascribable to treasury operations. Whereas in 1978 the increase in net credit to governments amounted to 11 percent of the total expansion in credit, in 1979 and 1980 it rose to 25 and 26 percent, respectively.

Data on the evolution of the composition of credit show that the rate of expansion of total domestic credit in the W AMU has accelerated since the revised statutes came into force in the mid-1970s (Tables 23-25 in the Statistical Appendix). The rate of expansion of credit to the government sector also exceeded that of credit to the private sector. This trend has been most apparent since 1978, when credit to governments rose sharply in all countries except Senegal (where the sudden increase came in 1979), while credit to the private sector rose more moderately. This result is more pronounced in Ivory Coast, Niger, and Senegal. The expansion of credit in each country also shows a fair degree of instability. For all WAMU countries, too, aggregate credit shows an explosive expansion after 1973, but the expansion is smooth, suggesting compensatory credit movements between member countries. This is likely, however, to have been fortuitous rather than reflecting a deliberate central policy decision.

    Other Resources Citing This Publication