Issues of income distribution, unemployment, housing, and social welfare have always loomed large in Algeria. With a young and rapidly growing population, the labor force has been increasing faster than job opportunities. The consequent unemployment, coupled with rapid urbanization and an existing housing shortage, has placed a severe strain on the social fabric and the provision of basic social services. Nevertheless, Algeria entered the 1990s with less acute poverty and inequalities in income distribution than other countries in the region with similar income levels.
The 1994 reform program had important implications in this regard. Prior to the reform, the government had tried to provide a social safety net via generalized subsidies, extensive and sheltered employment in the public sector, and income transfers. This system was both inequitable and inefficient, and, with growing fiscal imbalances, it also became financially unsustainable. Measures that had to be undertaken to liberalize the economy—trade and price liberalization, public sector enterprise restructuring and privatization, financial sector reform, and the overhaul of the housing delivery system—as well as the sharp decline in real wages entailed by the adjustment process, shook the existing paradigm of social protection and rendered its transformation unavoidable.
It was clear that an improvement in living conditions was essential to gain the support of the population for the reform program, especially in a difficult political and social environment. Therefore, the program included measures aimed at mitigating the transitional costs of structural adjustment for the most vulnerable sectors of the population, mainly through the overhaul of the social safety net, including the replacement of the previous system of low and poorly targeted income transfers by a self-targeted public works program, and the establishment of an unemployment insurance scheme.
Ultimately, the success of reforms will be measured by the ability of the Algerian economy to bring about a sustained improvement in the welfare of the population. Thus, over the medium term, a key challenge for Algeria will be its ability to provide adequate education, health, infrastructure, and housing services; to create sufficient employment opportunities to reduce the existing unemployment and absorb new entrants; and to protect the most vulnerable sectors of the population. This section seeks to identify the main problems underlying these challenges and reviews the strategies that are being pursued to address them.
Algeria’s population in 1995 was estimated at 29.6 million, of which more than one-half were under the age of 20. While population growth has been steadily slowing in the recent decade, to an estimated 2.1 percent in 1995, the higher rates in the 1970s and 1980s explain the young age structure of the population. As a result of this age structure, the labor force has been growing much faster than the population, at a rate of nearly 4 percent a year in 1981–95.20 Total labor force, however, stood at an estimated 7.1 million in 1995, only 25 percent of the population, owing to the large school enrollment and low participation of women compared with other countries in the region. Nevertheless, labor force participation has been rising in recent years.21 In particular, the marked increase in the female participation rate—despite increased schooling, which led to declining participation rates for younger women—may indicate a response to falling real wages and large family size.
Since the pace of employment creation in the past decade was insufficient to absorb the large number of new entrants in the labor force, unemployment escalated from 10 percent in 1985 to 25 percent in 1995, when an estimated 5.3 million Algerians were employed, and 1.8 million unemployed.22 Thus, on average, the earnings of one employed person was needed to support five people. Among the employed, self-employment remains substantial, probably a reflection of a large service sector and the importance of agriculture. The structure of unemployment also remains quite skewed (see Box 8).
Dynamics of Unemployment
Labor market prospects remain extremely clouded, a matter of great concern to Algerian policymakers.23 With the tightening of further migration to western Europe, the future evolution of labor supply in Algeria will be mainly determined by two elements: (1) the evolution of the population aged 15 and over and (2) the likely evolution of the labor force participation rate. Regarding the latter, one would expect male labor force participation to accelerate as the fraction of men aged 12 to 25 years and attending school is reduced; female labor force participation is also likely to continue to rise, as an increasing proportion of women will take advantage of their education to join the labor market and raise household income. Under the assumption that participation (as a percentage of the population) continues to rise through 2010 in line with recent trends, while population growth continues to slow down, about 250,000 people would be added on a net basis to the Algerian labor market annually.
Appendix II presents some unemployment projections under a “low-growth” scenario and a “high-growth” scenario. Under the low-growth scenario, and assuming high elasticities of employment/output creation, unemployment would fall to 26 percent by 2001 and 24 percent by 2010. Making more pessimistic assumptions on elasticities, unemployment would rise rapidly to about 31 percent by 2001 and 37 percent by 2010. Under the high-growth scenario, and assuming high elasticities, unemployment would fall to 23 percent by 2001 and be reduced to about 8 percent by 2010. Assuming low elasticities, high growth would be insufficient to reduce unemployment, which would gradually rise to 32 percent by 2010. These projections underscore the need for growth to be not only high and sustainable, but also more labor-intensive than in the past.
Reasons for Unemployment
Several explanations may be advanced to account for the high level of unemployment. First, the growth rate of real non-oil GDP between 1985–95 has been substantially below the rate of increase of the labor force and the increase in the labor participation rate. Second, major distortions in the price system and an overvaluation of the exchange rate that prevailed until 1994 favored capital-intensive technology. Third, the adjustment process has resulted in some labor shedding, whereas the participation rate has increased, particularly among women. Fourth, there are distortions embodied in certain features of the institutional structure of the Algerian labor market and labor regulations that may have reduced employment in the formal sector while raising it in the informal sector. Finally, educational and geographical mismatches also play an important role.
Insufficient job creation in private sector. About 1 million job seekers entered the market during 1991–95, while the economy created only 0.4 million new jobs—virtually all in the public sector (0.1 million) and in commerce and services (0.3 million); employment in agriculture, industry, and construction broadly stagnated. Similar trends had been observed during 1985–91. Important differences were also observable in private and public sector hiring: annual growth rates of employment during 1985–89 in privately owned industrial and construction firms averaged 6 percent and 10 percent, respectively, albeit from a small base. These trends are likely to have accelerated since.
Trade unions, minimum wages, and wage bargaining. The legally binding minimum wage (salaire national minimum garanti, SNMG) is set by the “tripartite”—the government in consultation with labor unions (the Union Generate des Travailleurs Algeriens, which until 1990 had a statutory monopoly) and employers (both public and private) at the national level. Additional agreements (conventions collectives) are struck at the sectoral level, within the limits set out by national settlements. Civil service salaries are determined according to a grid that, in the past, was also valid for public enterprises. Following the introduction of the new labor code in 1990, public enterprises were granted more autonomy in wage setting. Civil service salaries, however, still served as a reference point. In the past, both increases in the minimum wage and the average wage were negotiated for the public sector. For instance, in 1994, the agreed salary increase applied to all public enterprises, even those that were shedding labor under restructuring plans, and where workers had accepted to trade salary increases against job retention. This policy caused major financial problems for some enterprises and, in 1996, the policy was abandoned. Thereafter, only increases in the minimum wage are decided at the national level.
Evolution of wages. The institutional setup of the labor market has likely contributed to upward pressures on wages. Data on minimum wages and public sector wages provide some indication of union pressures. The minimum wage rose in real terms by about 17 percent during 1989–93, while total labor productivity fell by 12 percent; the real minimum wage subsequently fell sharply during 1994–97, while productivity increased. Indicators of compensation levels in the private sector suggest that wages for the unskilled are substantially higher than the SNMG. In addition, the bias against labor—through the subsidization of capital24—in the 1980s was tackled by the sharp Algerian dinar devaluation and the increase in interest rates to positive real levels following the 1994 reform. Government wages also have been falling in real terms since 1993, and by 10 percent during 1995 alone. While remaining well above the SNMG, a more compressed wage distribution is emerging in the public sector. Wage levels in the private sector are thought to be lower for low-skill jobs, but more dispersed at higher-skill levels. The much larger share of the wage bill in total output observed in the public sector, compared to the private sector, suggests that union pressures, in part supported by extensive job protection legislation, have raised the cost of employment in the formal sector. In addition, many educated job seekers appear to prefer employment in the public sector, further stifling the development of a productive formal private sector.
Job protection regulation. Before the reform of the labor code in 1990, dismissal of workers was only possible in severe disciplinary cases. The new law permits labor shedding for economic reasons, but only after collective negotiations, arbitrated by labor inspectors, have failed to find an alternative solution. After shedding labor, employers are not allowed to take on new recruits at similar grades. Through 1994, dismissed workers were entitled to severance pay equivalent to one month’s pay for each year of service up to a maximum of 15 months. Despite far-reaching reforms introduced since 1994, including the over-haul of the severance payments and unemployment insurance system (see below), the current regulations still contribute to Algeria’s high un-employment: by creating incentives for insiders to ask for higher-than-market clearing wages; and by raising the cost of dismissal, thereby inducing employers to hire low-risk workers with a good work history, thus limiting the prospects for the young to find a job in the formal sector.
Large informal sector.25 Job protection legislation, minimum wage legislation, and sectoral or national wage agreements only cover the formal sector, and their existence has contributed to the development of the informal sector. In 1991, for example, only an estimated 27 percent of all private sector workers were salaried employees covered by the minimum wage. Labor market legislation and official wage settlements are likely to be important for the informal sector as well. As most informal activities take place within sectors of relatively low productivity (e.g., services), the availability of a few secure and well-paid jobs in the formal sector may raise unemployment by inducing workers to “line up” for them. For instance, university graduates may prefer to wait for a “lifetime” job in the public sector.
Educational and geographical mismatches. Algeria suffers from an educational mismatch, with too many school graduates trained in the social sciences and thus better suited to join the public sector. Private sector needs have not been taken into account in the education system, and a major reorientation of labor training is required. More-over, job intermediation continues to be inefficient. Only since 1990 have employers been free to hire as they choose, and all vacancies need to be registered with the public employment agency (Agence Nationale de l’Emploi (ANEM)). The role of ANEM has declined in recent years, and firms have increasingly developed their own recruitment channels. Geographical mismatches are also prevalent: according to a survey from the early 1990s, one-fourth of the unemployed still would not move to another province to get a job, although the share of unemployed who would accept a job of lower qualification, or in another sector or region, has increased somewhat since 1989. In addition, the housing shortage has imposed a severe constraint on labor mobility.
Labor Market Reform
Structural reforms of the labor market are an essential component of any employment-generating growth strategy. These reforms need in particular to promote employment in the formal sector because of its higher productivity and focus on the production of tradable goods. A reform agenda aimed at lowering unemployment and fostering employment in the formal sector should include the following elements:
First, further reform of job protection legislation is needed to facilitate labor shedding, provided sufficient advance notice is given. The cost of labor shedding for employers should be reduced by lowering their contribution to severance pay and by relying more on the recently introduced unemployment insurance scheme under which employers and workers would pool costs and risks. Compensation under this scheme should be limited to a shorter time period to foster an intensive job search and avoid subsidizing long-term unemployment.
Second, the legal minimum wage should serve only to protect the weaker segments in the population and not as a benchmark for the wage structure. The recent delinking of the civil service wage increases from wage increases in public enterprises, with the latter being determined in light of their individual financial situation, is a step in the right direction. A prudent minimum wage policy is all the more important, as payroll taxes in the formal sector are substantial.26 Such taxes further exacerbate the shift of unskilled workers into the informal sector.
Third, measures are needed to reduce the high youth unemployment rate and foster employment of the young in the formal sector. The applicability of all job protection and minimum wage legislation could be limited to individuals older than 30 years. As a first step, the 1989 Finance Law introduced a series of exemptions to wage taxes applicable to the employment of youth up to age 30. This could be complemented by removing the minimum wage requirement for a period of apprenticeship.
Fourth, reforms in the educational system are needed, so as to adjust the skill profile toward the private sector. For example, labor-market entrants need to be provided with more market-relevant, technical skills, and formal apprenticeship schemes.27
Further reforms of the labor market and a continued restrained incomes policy to make economic growth more labor intensive would render more likely the “high-elasticity outcomes” discussed above. Robust growth in the construction sector to alleviate the housing shortage, in particular, could contribute substantially to further raising the labor intensity of GDP. Strong emphasis on labor-intensive activities (e.g., housing and agriculture) by providing the necessary institutional support would help raise employment elasticities with respect to investment over time. At the same time, strong and credible reform policies may help Algeria to achieve growth well above 6 percent, which even in the presence of relatively low elasticities would help generate faster employment gains.
Overhauling the Social Safety Net
The 1994 reform program involved substantial modifications to the safety net, with the introduction of two new elements: (1) a self-targeted public works program to substitute for the generalized subsidies, which were gradually being eliminated, and for the transfers to persons without income, which were not well targeted; and (2) an unemployment insurance system to facilitate industrial restructuring. At the same time, the government started incorporating explicitly in the budget the implicit subsidies to the housing sector.
Generalized food and energy subsidies were costly, wasteful, and inequitable, since they were disproportionately benefiting higher income groups. First, generalized subsidies were expensive, amounting to 4 percent of GDP in 1990.28 In addition, energy subsidies, in the form of forgone profit transfers to the treasury from the oil production and refining companies, amounted to another 4 percent of GDP. Second, the system was inefficient inasmuch as there were large leakages of income transfers to the middle class. Moreover, subsidized products were diverted toward unintended uses, and those subsidized goods that could easily be transported (e.g., powdered milk, semolina, and petroleum products) were smuggled to neighboring countries.
The expenditure of the Compensation Fund rose sharply as a result of the Algerian dinar depreciation during 1990 and, in early 1991, the authorities introduced several measures to reduce the number, scope, and extent of subsidies (see Box 9). The subsidy system, however, continued to involve distortions and large costs to the government and remained poorly targeted.29 These distortions and the need to reduce the budget deficit set the stage for a further reduction in subsidies. In June 1992, administered prices for food items were increased. As a result, the budgetary cost of food subsidies declined from the equivalent of 4 percent of GDP in 1990 to 2.3 percent of GDP in 1992. Subsequent increases in the import cost of these commodities raised subsidies to 2.5 percent of GDP in 1993, because of the lack of a pass-through mechanism for administered prices.
Structure of Unemployment
In Algeria, as in most North African countries, the unemployed are mainly under 25 years old, typically remain unemployed for more than one year, mostly live in urban areas, and tend to have at least a secondary education. Moreover, female unemployment is wide-spread. This points to labor market rigidities that would need to be addressed.
High urban unemployment. The proportion of labor force participants that are self-employed has been rising steadily, accounting for one-fourth of all employed in 199.1. As the self-employed almost by definition cannot become unemployed, the aggregate unemployment rate reflects pressures on labor markets only imperfectly. As self-employment and unpaid family employment are particularly large in agriculture, it is estimated that urban unemployment rates are even higher than the national average.
Geographical mismatches. For both social and economic reasons, geographical mismatches are prevalent. People are generally reluctant to move to another province to get a job, and their ability to do so remains highly constrained by the acute housing shortage.
High youth unemployment. The unemployment rate among the young is nearly twice the global rate, in line with the situation in many Mediterranean countries. In 1991, about 41 percent of the labor force aged 20–25 were unemployed. The likelihood of being unemployed falls rapidly with age, and among the labor force aged 30 and older, only 6 percent were unemployed. Given the large share of youth in the total labor force, this implies that the bulk of the unemployed are under 25 years old, raising major social and political concerns, in part because of the weakness of the social safety net.
Educational mismatches. Unemployment data broken down by educational background indicate that unemployment rates do not necessarily decline with the amount of education received. Indeed, the unemployment rate of individuals with a middle or secondary education is highest (26–30 percent in 1991). The un-employment rate is lowest among those with postsecondary education (6 percent) and those with no education (14 percent). People with primary school education may be more successful in finding the jobs of their choice. Primary school degree holders are likely to work within the informal sector, as self- or home-employed, while those with a secondary school degree line up for the few formal sector jobs, notably the secure jobs in public administration. As recruitment in the public sector slowed down during the second half of the 1980s, graduates had to wait longer before being offered a job. Furthermore, a large share of graduates may have had degrees in the social sciences, and their job opportunities in the private sector were therefore more limited.
Long unemployment duration. Recent data on the proportion of long-term unemployed—for more than one year—in total unemployment are not available. Based on labor surveys, the average length of unemployment spells was 23 months in 1991; by age, the average length of unemployment peaked for the age group 25–34 at 36 months, and was longer for men than for women. The long-term unemployed are likely to be much younger in Algeria.
To shield the poor from the 1992 adjustment to administered prices for food, the authorities introduced in February 1992 a system of transfers that provided cash allowances to four types of beneficiaries: low-income earners and their children, pensioners, and persons without income. By 1993, this scheme covered more than 60 percent of the population and cost about 2 percent of GDP. Its main weakness was that it was poorly targeted due to inadequate means-testing. Indeed, for low-income earners and their children and pensioners, targeting was based solely on the records of the Social Security Administration. Targeting was also deficient in the case of persons without income. Beneficiaries were only required to register their need and to “declare on their honor” that they did not have any source of income; ad hoc verifications would be carried out in certain cases, using the records of various agencies, such as the Tax Department, the Social Security Administration, or the Chamber of Commerce. Because a large number of informal sector activities and intrahousehold transfers were not captured, these attempts at targeting proved inadequate with the result that 25 percent of the population were considered “without income” and benefited from cash transfers.
Beginning in 1994, cash compensation and the remaining generalized subsidies were gradually replaced by targeted transfers and subsidies. By the end of 1996, the phase-out of generalized food subsidies was completed. By the end of 1997, the overall subsidy on gas and electricity was completely eliminated, through quarterly increases in their prices (see Box 9).
Public Works Program
To partially substitute for both the generalized subsidies, which were gradually being eliminated, and the allowance paid to persons without income, which was not well targeted, a self-targeted public works program, managed at the local level, was introduced in October 1994. It was designed to provide compensation to those able to work (Indemnité d’activités d’intéret général—IAIG) and to provide financial support to those unable to work (Allocation forfaitaire de solidarity—AFS). Those willing to work can be hired at about one-half the minimum wage—in community-based activities, such as reforestation, water works, and street cleaning. The low remuneration is designed to ensure self-targeting of those with a low opportunity cost. Those unable to work—pensioners and the disabled—receive DA 900 a month, plus DA 120 for each dependent up to a maximum of three. The public works program appeared most appropriate under the circumstances, given the pressing problem of youth unemployment and the previous un-successful experience with targeting.30 On account of improved targeting, benefits to those eligible under the new system are higher than under the old system.31 Because Algeria had experimented with community-based public works in the past, these activities could be designed and implemented quickly.32
A recently conducted survey indicates that the public works program has been quite successful. By the fourth quarter of 1996, there were 500,000 participants under the scheme, working for an average of eight weeks. In addition, 277,000 elderly and disabled persons participated in the cash compensation scheme, bringing the total number of participants to 7 percent of the population, and indicating that the scheme operates under substantially improved targeting. In 1997, the IAIG was supplemented by a more structured public works program supported by the World Bank and aimed at improving rural infrastructure through the creation of 20,000 yearly jobs.
Reform of Unemployment Insurance
A major reform of the unemployment insurance system was implemented in July 1994. Before the reform, enterprises were required to provide severance payments of one month of salary for each year of service to retrenched workers, paid as a lump sum at the time of layoff. Although not particularly high compared to other countries, these severance obligations in cash were onerous for enterprises in financial distress and acted as an impediment to restruc-turing and labor shedding.
Under the new system, retrenched workers receive a lump sum severance payment limited to three months and are not eligible for unemployment benefits during that time, to encourage active job search. Employers have to pay an “entry fee” to the Unemployment Insurance Fund (equal to 0.8 month of salary for each year of service of the retrenched worker in excess of three years, up to a maximum of 12 months) at the time of layoff. The new scheme alleviates the burden of effecting layoffs in two ways. First, it lowers the amount of severance pay by reducing compensation for employees with 3–15 years of service.33 Second, the entry fee can be paid in monthly installments over one year. An additional advantage of the new scheme is that the small lumpsum payment and ineligibility to benefits during the first three months of unemployment encourage search behavior; it also encourages the creation of small enterprises, the most dynamic area of the Algerian private sector.
In addition to entry rights, the Unemployment Insurance Fund collects a monthly unemployment insurance contribution of 1.5 percent of the salary base from employees and 2.5 percent from employers. These payroll contributions serve to finance monthly unemployment benefits to which retrenched workers become entitled after three months. Unemployment benefits are based on a reference salary of the average of the gross monthly salary and the minimum wage, the SNMG. The duration of benefits equals two months for each year of contribution to the Unemployment Insurance Fund while at the last employer, with a maximum of 36 months. The eligible period for benefits is divided equally into four periods, with benefits declining gradually over these periods: from 100 percent of the reference salary in the first period, to 80 percent in the second period, 60 percent in the third period, and 50 percent in the fourth period (subject to a minimum benefit of 75 percent of SNMG and a maximum of three times the SNMG).34 World Bank simulations show that the new unemployment insurance system would be financially viable even at current un-employment rates.
Reform of the Price System and Elimination of Subsidies
Prior to 1994, the domestic price system in Algeria was subject to numerous controls and distortions. The price law of 1989 divided products into (1) those with administered prices;1 (2) those subject to controlled profit margins;2 and (3) those for which prices were free but still needed to be declared to the authorities. Generalized subsidies were also provided to consumers for energy products and 15 major food staples; and to producers of several agricultural products and inputs. Aside from the misallocation of resources, which this system entailed, it was also inequitable since the richer half of the population were consuming more than 60 percent of the subsidized food items. Since 1994, domestic prices have been almost entirely liberalized, with most controls on prices and profit margins lifted, and most subsidies eliminated.
Lifting of controls on prices and profit margins.A major step was taken in April 1994, through the abolition of controls on profit margins for most commodities.3 Price controls remained only for three subsidized essential food staples (flour, semolina, and milk), energy products, and public transportation fares. Later in 1994, prices of agricultural inputs and construction prices for social housing were liberalized, and all remaining controls on profit margins4 and prices were eliminated during the 1995/96 program year (April 1, 1995-March 31, 1996), except for medicines and subsidized food and energy products.
Elimination of generalized subsidies. Reform of the domestic price system also entailed major increases in the prices of subsidized products toward their opportunity costs. Prices of subsidized food and energy products were doubled in 1994/95 and increased by 60 percent in 1995/96. As a result of these price adjustments, subsidies were eliminated on petroleum products (October 1994), powdered milk and semolina (June 1995), regular flour (October 1995), and bread flour (January 1996). By the end of 1996, all food subsidies had been phased out. For energy products, the implicit subsidy was eliminated by setting the transfer price of crude oil from Sonatrach to the refineries at world price level, with adjustments to the transfer price every six months in line with export prices and exchange rate developments. There also remains a global subsidy on gas and electricity products (equivalent to 0.4 percent of GDP in 1996), which the government is committed to eliminate by the end of 1997. The system of quarterly revisions of the prices of electricity and gas introduced in August 1994 should prevent the reemergence of this global subsidy after the end of 1997. Nevertheless, subsidies of gas and electricity to poor households—which can be targeted through consumption levels—will continue.
Adoption of a Competition Law in January 1995, institutionalizing the principle of free price setting for all products while introducing safeguards against possible abuses by monopolistic suppliers, including through antitrust regulations. The law prohibits noncompetitive practices, such as the imposition of limits on market access or collusion between enterprises to control a particular market. This regulatory framework is being enforced by a number of institutions, including a National Competition Council, the Ministry of Commerce, and the Ministry of Justice.1 Beyond explicitly subsidized items, this category included cereals, water, gas, electricity, fuel, transportation fares, harbor services, construction costs, and cement.2 This category included tea, coffee, milk, metal cans for food preservation, plastic sheets for agricultural use, agricultural equipment and parts, fertilizers, lubricants, yeast, cattle feed, notebooks, textbooks and school furniture, medicines, powdered sugar, edible oils, tomato concentrate, medical instruments, furniture, equipment and materials, tobacco products and matches, semolina, high-grade flour, and pasta.3 Controls on profit margins were eliminated in mid-1995 for sugar, cereal grains (other than hard and soft wheat), and edible oils, and at the end of 1995 for textbooks and other school supplies.4 Except for edible oils, sugar, medicines, school supplies, coffee and tobacco, and five cereal products that were trans-ferred from the category of goods with administered prices to that of goods with controlled profit margins.
Housing Reform: Fostering a Market-Driven Supply Response
Housing needs have increased considerably in Algeria since independence in 1962, mainly as a result of rapid population growth and a high pace of urbanization (see Box 10). Housing supply has lagged behind demand, as the housing sector remained overwhelmingly dominated by inefficient state-owned construction enterprises, public real estate management companies, and a state-owned housing financial institution (CNEP). Long and costly delays for housing completion plagued the sector, particularly in the early 1990s, when imports of construction inputs were curtailed as a result of foreign exchange rationing. Provision of subsidized public housing covered a broad income group and was mostly given out to public sector employees. While the deposit base of the CNEP was channeled to finance public rental housing, financing for unsubsidized private housing was virtually nonexistent. As a result, Algeria has a severe housing shortage that has largely contributed to social discontent in recent years. With a housing stock of low average quality and of less than 4 million units for a population of more than 28 million, it has one of the highest occupancy ratios in the world (Figure 8).
Housing supply in Algeria has fallen short of the needs of a rapidly increasing population in Algeria, giving rise to a serious social problem, whose solution remains an important government priority. The housing stock is estimated at about 4 million units for a population of about 29 million, one of the highest occupancy ratios of the world. Moreover, the housing stock is of poor quality, with half of the dwellings over 35 years old. This, coupled with below market rents, has exerted increasing pressures on the budget. The housing shortage has stemmed essentially from the combination of a dysfunctional and cumbersome state-organized market with inefficient state-owned construction companies, poorly managed real state companies (OPGIs), and a public savings and housing financial institution unable to play properly its intermediation role.
Over the period covered by the program supported by the extended arrangement, the government took several steps to streamline its interventions, liberalize the sector, and ensure greater efficiency in the construction of public housing. In the early stages of reform, the government—in an attempt to establish market mechanisms and eliminate distortions—lifted trade and payment restrictions on imported inputs and liberalized the price of construction. At the same time, to minimize budgetary costs, the government started adjusting rents gradually, with the goal of reflecting construction costs better. However, even after sharp increases, including 30 percent in 1997 and again about 20 percent expected for 1998, rents in public housing remain well below market prices. Nonetheless, since January 1998, all new units delivered have been rented at a higher rate.
The government has initiated a progressive withdrawal from direct involvement in construction and taken actions to improve efficiency and reduce costs in its interventions in the provision of social housing. As a result, a large number of state-owned construction companies have been liquidated with about 80,000 workers dismissed as of December 1997. Meanwhile, the number of property management and promotion agencies (OPGIs) was progressively reduced from 53 in 1996 to 43 at the end of 1997, and is to be reduced to 25 by the end of 1998. In addition, in 1997, a Cost Reduction Commission was created to ensure that public projects are carried out at minimum cost. An assistance program has been elaborated to improve the targeting of social housing; however, the need to collect reliable data on household income has delayed the implementation of this mechanism. Finally, the allocation of public housing is now the full responsibility of the communes (instead of the OPGIs), which can judge better the needs of the local population, hence leading to a more just allocation of social housing.
The financial framework in place was a mayor impediment to help develop a private construction sector and, thus, stimulate the adequate supply response for the massive shortfall in housing. In response, the government began a far-reaching overhaul of the mechanisms for housing financing. In 1997, the Caisse nationale d’epargne et de prevoyance (CNEP) acquired the status of a bank and was recapitalized so as to revitalize its role of intermediary in the sector, given that it mobilizes a significant portion of household savings. At the same time, new auxiliary financial institutions were created to open housing finance to the entire banking sector and to better control related risks. To this end, a mortgage refinancing company (“Societe de refinancement hypothecaire”), a mortgage guarantee company (“Societe de garantie du credit immobilier”), and a fund to provide guarantees to housing companies (“Fonds de garantie et de caution mutuelle de la promotion immobiliere”) were created. These institutions are expected to become fully functional in the course of 1998, after the operating mechanisms have been put into place with foreign technical assistance. The capital of these new institutions was provided essentially by the banking sector.
The 1997 government program called for an acceleration of the public construction program. Out of 142,000 units approved in 1995 under the previous social construction program (PEC), about 79,000 had been delivered by the end of 1997, and the remainder is to be delivered in 1998. In addition, the supplementary budget for 1998 envisages a new PEC of 80,000 units including 20,000 that were started at the end of 1997. About 5,000 of these new constructions are expected to be delivered in 1998, and construction will be performed by both private and public contractors. The contribution of the budget to public housing amounts to DA 80 billion (2.8 percent of GDP) in 1998. Also, a joint program with the World Bank to deal with the problem of shanty towns will be launched by the end of 1998 with a total contribution of $200 million.
Because of past centralized planning and rent control, the supply of housing has come mostly from the public sector. Almost 50 percent of the housing units delivered by the formal sector between 1980 and 1996 consisted of state-owned rental units. Housing supply has traditionally been heavily subsidized through a system of administered controls keeping construction prices, rents of public housing, and mortgage rates below market-clearing levels.
The system of implicit subsidization of housing resulted in mounting losses—financed by banks—for construction companies and public rental agencies, and produced only a limited supply of housing that has not reached the neediest segments of the population. Since 1994, this system has been gradually phased out and replaced by a system in which (1) subsidies on construction inputs and administered prices on construction costs have been eliminated; (2) most housing is supplied by the private sector at market prices; and (3) government subsidies for housing are explicitly budgeted and redirected to-ward those who otherwise could not afford housing. This change in strategy, which is expected to foster a substantial market-driven increase in housing supply, requires the restructuring of the housing construction sector, a reform of the rental housing market, and the revamping of housing finance.
Restructuring the Housing Construction Sector
Algeria had a large number of inefficient public construction enterprises that employed about 240,000 workers in 1994 and accounted for more than 80 percent of the total value added of the construction sector. The restructuring of the sector was initiated in 1992 with the support of the World Bank Enterprise and Financial Sector Adjustment Loan, and was accelerated first in 1994 with the liberalization of imports of construction materials and construction prices, and since 1995, through various measures aiming at both hardening budget constraints faced by public enterprises and increasing participation of the private sector in the supply of housing. In particular, public construction companies have been granted autonomy while nonviable ones are being liquidated. By 1997, numerous public construction companies had been liquidated. The workforce of public construction enterprises has been reduced and most labor contracts were trans-formed from full-time employment to contractual employment. In this context, about 93,000 workers were laid off in 1995–97, or 51 percent of the current employment in public construction companies. Finally, public enterprises have increased recourse to private subcontractors in the finishing stages of construction. To this end, contracts for public housing projects were parceled out into small components, more in line with the capacity of private contractors.
As a result of these actions, the participation of the private sector in the construction of public housing for rent has increased from 20 percent in 1994 to more than 50 percent in 1997. The number of formal construction projects for sale launched by the private sector is estimated to have increased by more than 50 percent since 1994. The increased involvement of small private construction firms has also contributed to a decline in the completion period for more than five years to one year for some projects, which, in turn, has translated into a substantial fall in construction costs.
Overall, yearly housing deliveries by the formal sector are estimated to have increased from 82,000 units in 1994 to 130,000 units in 1996 (Figure 9). The increase mostly occurred in public housing and stemmed from efforts at finishing dwellings launched prior to the 1994 adjustment program. The stock of ongoing construction projects at the end of 1996 was estimated at 170,000 units, including 110,000 units of public housing for rent. The latter could be completed in about two years assuming that the pace of housing deliveries is maintained at the level achieved in 1995 and 1996. As part of the shift from rental public housing to the promotion of owner-occupied housing, however, these completed units will be sold.
Over the next few years, restructuring of the housing sector will require an acceleration of the liquidation of nonviable public construction enterprises and the privatization of others. As shown by the recent experience of the transition economies of eastern Europe, privatizing housing construction can quickly lead to significant productivity gains as this activity is particularly suited for small- to medium-scale enterprises. Many aspects of housing construction are relatively labor-intensive and can be easily subcontracted to small enterprises organized into specialized guilds. The possibility of privatization in the construction sector is, therefore, not as dependent on the availability of foreign direct investment as that of other more capital-intensive sectors in the Algerian economy.
Reforming the Housing Rental Market
The supply of rental housing in Algeria has been almost exclusively provided by the government through public rental companies (OPGIs) at rent well below market-clearing levels. At the end of 1996, rents for a publicly owned, two-bedroom city apartment, ranged from $8 to $12, while private sector rents for similar dwellings ranged from $55 to $85. In 1996, rents actually collected by the OPGIs were still only about 40 percent of rents due, and covered less than 70 percent of personnel and maintenance costs. Moreover, public rental housing was made accessible to a broad spectrum of middle-income tenants, including a large segment of public sector employees.
Several measures were implemented in 1996 to remedy the structural causes of financial imbalances of the OPGIs. Specifically, the Civil and Penal Codes were amended with a view to placing the burden of proof of rental payments on tenants and allowing the licensing of private property managers. In the context of this new legal framework, the government, in 1997/98, will proceed to restructure viable OPGIs and liquidate loss-making ones. The restructuring process will entail the privatization of the existing stock of public housing for rent as well as the subcontracting of its maintenance to the private sector. As a result of the transfer of these functions to the private sector, remaining OPGIs will only retain a limited staff to perform their role as contractors for public housing programs.
Concurrently, subsidized public rents were increased substantially between March 1994 and December 1997. The idea is to gradually raise rents for public housing to market levels, while putting in place a system of well-targeted and fully budgetized direct subsidies to low-income tenants. During the transition period, any loss made by OPGIs will be accounted for in the government budget. These measures should facilitate the purchase of state-owned apartments by their tenants and contribute to the development of a private rental market for housing.
Revamping Housing Finance
Housing finance in Algeria has de facto been monopolized by the state-owned housing fund (CNEP). Although commercial banks are legally allowed to extend mortgage financing to households, the CNEP was the only financial institution providing mortgage loans at preferential interest rates to holders of saving accounts. The number of concessional mortgage loans was limited by the availability of housing and benefited the better-off depositors who could meet the CNEP’s prescribed interest accumulation threshold. In addition, as noted earlier, most of the savings of the CNEP’s depositors were channeled to OPGIs to finance public housing for rent. The inability of OPGIs to service their debt with the CNEP has caused severe liquidity problems for the CNEP, particularly after 1995, when it depleted its stock of treasury bonds after having faced a stagnant depositor base under the old negative interest rate structure.
This situation has called for a major overhaul of housing finance on the basis of the following principles: (1) there must be a clear separation of the subsidization of housing and mortgage financing; (2) all banks should be in a position to participate, on a competitive basis, in the supply of mortgage loans; and (3) mortgage loans must be supported by clearly defined property rights to ensure that they can actually be recovered.
With the first principle, the government has decided to channel all public subsidies to housing as well as finance any new public housing project exclusively through the budget. The government also restructured the CNEP into a housing bank operating solely on a commercial basis in early 1997. To this end, in April 1997, government recapitalized the CNEP and took over all the OPGI’s nonperformance debt to the CNEP in exchange for government bonds. Interest rates on CNEP deposits were increased significantly over the last two years, resulting in a substantial growth of deposits in 1996–97.
To facilitate the broadening of mortgage financing to all commercial banks, a mortgage refinancing institution was created in 1997. This new institution will deal only with other financial institutions and will allow commercial banks to participate in the mortgage market while keeping the maturity structure of their assets in line with that of their liabilities.
Finally, regarding property rights, a major operation has been launched since 1993 to update and extend the coverage of the national land registry. This process, which is expected to be completed by 2008, should be accelerated to strengthen the legal basis for property registration. In addition, the availability of land for construction purposes will be increased as a result of a land privatization law, which should be enacted in 1998.