Brazil's economy grew rapidly until 1980, when expansion was brought to a halt by hyperinflation and the resulting instability. The country's living standards have been stagnant ever since (Exhibit 1). Yet disappointing though the performance of recent years has been, Brazil has tremendous potential for growth. While the government estimates that the economy could grow by 5 to 7 percent a year, our research suggests that if economic reform were strenuously pursued, growth could reach 8.5 percent a year.1 Progress of this order would lead to a doubling of per capita gross domestic product to about $10,000 within 10 years.
To examine the productivity and expansion of Brazil's economy, we carried out case studies of eight important industries: airlines, automotive, food processing, food retailing, residential construction, retail banking, steel, and telecommunications. We then used our findings to estimate what growth rates might be like if there were extensive reform of the policies and regulations governing these industries.
Our main findings were:
-
Rapid productivity increases and additional domestic and foreign investment are the keys to doubling Brazil's per capita GDP.
-
To achieve these gains, Brazil must hold inflation in check and ensure macroeconomic stability. Deregulation, reductions in import tariffs, and privatization must also proceed in order to increase competition and exposure to global best practices.
-
Investment must rise to 26 percent of GDP from its current rate of 19 percent. This could be achieved through productivity growth, by moving toward a balanced budget, by streamlining the banking system, and by introducing further incentives for private savings. With sound macroeconomic management, the economy could support foreign direct investment and foreign indebtedness to the level of half of Brazil's GDP without over-reliance on outside capital.
-
Brazil could double its GDP per capita without first having to attain the educational level of countries that already have double its GDP per capita by instituting company training programs and global best practices.
-
This fast growth path will not generate structural unemployment. On the contrary, fast growth is essential if Brazil is to make any progress toward reducing poverty.
Productivity: The engine of growth
For any country, the quickest and surest way to raise living standards is to raise productivity. The gap between Brazil's GDP per capita and that of more developed economies arises largely from low labor productivity and capital inputs (Exhibit 2).
Exhibit 3 compares Brazil's labor productivity in the industries we studied with that of the United States, and shows how employment is distributed among these industries. With the exception of steel, productivity is less than half the US level, even in modernized industries such as airlines, telecommunications, retail banking, and auto assembly. In food processing and food retailing, productivity is less than 20 percent. Because a sector's contribution to a country's overall productivity is proportionate to the number of people it employs, raising productivity in the big domestic industries of residential construction, food processing, and food retailing is of greater importance than making gains in more modern sectors.
We tried to find out whether the reason productivity has not improved is that the low cost of labor deters investment in advanced technology. Capital in Brazil is far more expensive in relation to labor than it is in developed countries; during the 1980s, for example, capital was 16 times more expensive compared with labor than it was in the United States, according to our estimates. This figure has fallen recently, following a reduction in import tariffs and the implementation of the Real Plan, a program aimed at promoting macroeconomic stability. But it is still not always economically viable for Brazilian companies to automate as much as, say, the Japanese do in their steel plants.
Underinvestment in technology thus accounts for a small part of the productivity gap. But if we look more closely at operational processes, it becomes clear that a larger portion of the gap, about 35 percent, could be closed by improving the way work is organized, and another 30 percent by making modest capital investments with payback periods of less than two years. Simply to optimize existing processes without investing any new capital would increase GDP growth by almost 2 percentage points.
Large industries are still dominated by unproductive and highly fragmented "informal" workforces
Specific opportunities to close productivity gaps vary by sector. Large industries such as food retailing and construction are still dominated by unproductive and highly fragmented "informal" workforces: street vendors rather than department stores, individual handymen with toolboxes rather than construction companies. In food retailing, labor needs to be redeployed into more productive convenience formats along the lines of 7-Eleven in the United States and Japan, and Aldi in Germany. These companies use their scale to streamline store operations, purchasing, and logistics. In the automotive sector, the main areas needing improvement are factory layout (to reduce wasted effort when parts are moved between workstations) and quality control.
Progress toward the adoption of best practices was hindered during the 1980s and early 1990s by inflation. The timing of payments and price increases came to dominate companies' relationships with suppliers. Joint planning and just-in-time delivery, the cornerstones of world-class auto manufacturing, became virtually impossible. In industries with price controls, such as telecommunications, steel, food processing, and automobiles, managers had little incentive to control costs, since cost increases were the easiest way to justify price increases. Across the board, high inflation made it harder for consumers to compare prices, a state of affairs that reduced competitive intensity and with it managers' incentive to improve service and productivity.
Local markets were then, and often are still, stifled by regulatory limits on local and foreign competition. In the dairy and milling industries, for instance, a combination of price-fixing mechanisms and quota systems allowed even highly unproductive operators to survive. The resulting productivity levels, respectively 22 and 9 percent of US levels, contrast with the 68 percent achieved by the poultry industry, which is heavily focused on exports. Among airlines, regulation of domestic travel made the cost of a flight from São Paulo to Rio de Janeiro $300, compared with $80 for a flight of similar length in South Korea. Not surprisingly, five times as many Koreans travel by air as Brazilians.
The main product and capital market barriers yet to be addressed are tariffs and government ownership. Protective tariffs and lack of competition led to a $7.3 billion overspend on Brazil's telecommunications network, for example, and delayed the transition to digital technology, with its promise of higher productivity. Government ownership is perhaps most prevalent in banking, where private banks are almost twice as productive as public ones. As a result, $8 billion in extra banking costs are incurred each year.
It is in informal firms' interest to stay small and disorganized; growing larger would mean having to pay taxes
Productivity growth is further suppressed by inconsistent tax enforcement. In food retailing and house building, tax evasion by small informal companies reduces the cost advantages that larger tax-paying companies can obtain if they do improve productivity. It is in informal firms' interest to stay small and disorganized; growing larger would mean having to pay taxes. As a result, these businesses, which make up more than three-quarters of the construction and food retailing industries, are effectively locked into unproductive scale and levels of capital.
However, developments such as the product market deregulation of the early 1990s, the stabilization of the macroeconomic environment in 1994, and some recent privatizations have begun to dismantle these barriers to growth. If they continue to fall, Brazil has the potential to reach a productivity growth rate of 6 percent a year and sustain it for 10 years. Substantial restructuring is likely to take place in all sectors, ranging from privatization in telecommunications through the entry of new players in the automotive industry to consolidation in banking, food processing, food retailing, and construction. If productivity gains were combined with the investment needed to employ the extra workers from the projected annual increase in working-age population of 2.5 percent, the result would be a potential GDP growth rate of 8.5 percent.
This is an ambitious goal, but it is feasible. Starting from the position that Brazil occupies today, South Korea doubled its GDP per capita over 10 years by attaining productivity growth rates in many sectors similar to those we believe are possible in Brazil (Exhibit 4). And given that Brazilian best practice achieves productivity close to US levels, it is reasonable to believe a similar feat could be achieved in Brazil.
The feasibility of increased investment
We went on to investigate whether savings and investment would need to reach Asian levels to support an 8.5 percent growth rate, drawing on our recent research on South Korea.2 We concluded that Brazil could improve greatly on South Korea's achievements, and that it would not need to save and invest at South Korean levels to do so.
The reason is that the South Korean government's policy has been to invest large amounts of capital in selected sectors—capital that was then used inefficiently. This meant that to double its per capita GDP, South Korea needed investment of 33 percent of GDP. For Brazil to follow suit, its current investment rate of 19 percent of GDP would have to rise to no more than about 26 percent, although capacity use and throughput would have to improve. Our case studies suggest that such improvements can be made.
This 26 percent rate would put Brazil on a more productive development path than that of Japan and South Korea, but less productive than that of the United States (Exhibit 5). Capital productivity would then be much higher than South Korea's, although GDP per capita would be the same.
Higher productivity will itself generate much of the investment needed to increase capacity by lowering the price of investment goods and the cost of financial intermediation. But to achieve its full structural growth potential, Brazil must also increase its savings rate. The best way to do so is to balance the government budget. Higher investment will lead to increased imports of capital goods, a development we believe to be compatible with maintaining conditions for macroeconomic stability.
Training to achieve best practice
Brazilian workers have only 5.6 years of formal education, whereas workers in the United States, Japan, and South Korea have between 11 and 13. Nevertheless, our case studies indicate that with the help of corporate training, Brazilians could deliver substantial productivity improvements. In the food processing, food retailing, and automotive industries, companies are already able to raise productivity by compensating for their employees' lack of education with training, the overall cost of which is modest compared with its productivity benefits.
One large dairy in Minas Gerais has exceeded average US productivity; a leading volume retailer has reached 150 percent of average US productivity; and Honda's motorcycle plant in Manaus has reached productivity on a par with that at home. All have done so with their existing local workforces. Other companies planning to enter Brazil's automobile industry expect to reach home-country staffing and productivity levels within three to four years.
We found examples in the United States that support our belief in Brazilian workers' potential. A Houston house builder has recorded productivity four times higher than that of Brazilian counterparts using former agricultural workers from Mexico whose educational background is similar to that of the average Brazilian construction worker and who often do not speak fluent English. Similarly, a biscuit producer in Richmond, Virginia has invested in training to enable its employees, many of whom did not complete high school and have reading and writing difficulties, to perform complicated tasks in a highly automated plant. The secret of these best-practice companies is to invest in training workers for specific roles. In some cases, the training costs are higher for Brazilians than they are for more educated workers, but in no instance are they steep enough to cause a productivity penalty of more than 10 percent.
Since we did not systematically compare the educational backgrounds of workers across the companies in our case studies, we are not able to measure fully the impact of Brazilian educational standards on productivity. We can say, however, that apart from some oil-producing nations, every country that has achieved per capita income at least twice that of Brazil has a substantially more educated workforce than Brazil's. This suggests that government programs to give a larger proportion of Brazilian youth at least some secondary education would help to achieve the goal of doubling per capita income. Brazil will also have to increase the percentage of workers with college education from 16 percent to 20 percent.
Unemployment and poverty
Finally, we investigated whether productivity increases would necessarily lead to unemployment and increase the number of poor people. Our analysis showed that if growth were to reach 8.5 percent a year, subsistence and the informal economy would decline and there would be no increase in unemployment.
Our reasoning was that in competitive markets, productivity growth leads to lower prices (Exhibit 6), which in turn increase demand. Study of the food retailing indus-try, for example, indicates that food prices could fall by 40 percent if modern convenience channels were introduced. This would both increase demand for food and raise the disposable income of poorer people.
If the goal of doubling GDP per capita were indeed realized, a reallocation of labor between economic sectors would follow. Brazil has a flexible labor market with a low minimum wage and limited unemployment benefits. Every displaced worker thus has a strong incentive to find a new job. The result is low unemployment, even though Brazil has higher worker mobility than any developed country. With growth of 8.5 percent a year, employment would be redistributed from food retailing and food processing to house building, telecommunications, and airlines. This structural job displacement would represent less than 10 percent of normal worker mobility, leading us to expect that displaced workers would find new jobs. We also believe that rising demand for more skilled workers and better education would begin to redress wage inequality.
Brazil is poised to grow. If the government maintains macroeconomic stability, pursues deregulation and integration with the global economy, and removes barriers to productivity, we believe the economic security and well-being of Brazilians will rise rapidly over the next decade. 
About the Authors
Martin Baily is a principal, Vincent Palmade is a consultant, Eric Zitzewitz is a former consultant, and Bill Lewis is the director of the McKinsey Global Institute; Heinz-Peter Elstrodt is a director and Bill Jones is a consultant in McKinsey's São Paulo office; and Norbert Sack is a consultant in the Berlin office
Notes