41861 WORLD BANK CHINARESEARCH PAPER NO. 8 RAW MATERIALPRICES, WAGES,AND PROFITABILITYIN CHINA'S INDUSTRY--HOW WAS PROFITABILITY MAINTAINED WHEN INPUT PRICES AND WAGES INCREASED SO FAST? Song-Yi Kim */ and Louis Kuijs **/ October 2007 Abstract China's industrial sector has faced large increases in raw material prices, and wages have also risen significantly. This paper aims at analyzing and quantifying the impact of the cost pressures on the profitability of industry, as well as the way that China's industry has responded to the pressures. By combining information from the input output table and time series from the enterprise survey and on prices, wages, and employment, we estimate what has happened to the "technical efficiency" of the usage of intermediate inputs: the amount of intermediary inputs adjusted for price changes. We find that it has improved significantly since 2002 in most sectors in core manufacturing, although not for all. The second factor that has offset increases in input prices and wages is labor productivity growth, measured here as value added per worker. Our results suggest that sectors that faced higher cost pressures have made larger efforts to offset the cost pressures. The gap between output price rises and input price rises increased from 4 percentage points in 2005 to 4.3 percentage points in 2006 for core manufacturing. That this was possible while at the same time increasing the average profit margin shows that the ability of China's industry to offset rising raw material prices by increasing efficiency has so far remained undiminished. Keywords: China, profitability, commodity price, raw material, wage * The JFK School of Government, Harvard University. ** World Bank Office, Beijing, China; corresponding author; akuijs@worldbank.org. We thank Bert Hofman for encouragement and comments. The views expressed are those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. 1 I. Introduction Many have raised concerns about the impact of input price increases and wages in recent years and in the future on the profitability of China's industrial sector. However, to our knowledge, no systematic study has been done to quantify the impact of these cost pressures and to analyze quantitatively the way that China's industry has responded to them. This paper aims to fill this gap. Input prices have increased significantly since 2002. Many raw material prices rose very rapidly, even though for some primary commodities products China's prices did not grow as rapidly as international ones and prices of some "raw materials", including chemicals and certain machinery and equipment, rose only modestly. On average, raw material prices increased 34 percent between 2002 and 2006. As some rose much more in price than others, different sectors were hit differently, depending on which raw materials they consumed. While output prices (PPI) also rose significantly for most sectors, they generally lagged behind raw material prices significantly. As a result, relative (or, real) input prices went up substantially for almost all sectors, putting pressure on profitability. Wages rose even faster, adding to the pressure. Yet, profit margins did not decline in most industrial sectors. On the contrary, they have continued to rise in most sectors. In this paper we investigate how these developments add up. Using information from China's (2002) input output table and time series from the enterprise survey and on prices, wages, employment, we quantify the costs pressures bearing on different sectors. We also look at what have been the key avenues by which sectors have offset the cost pressures. By putting together this information and data, we are able to estimate what has happened to the "technical efficiency" of the usage of intermediate inputs: the amount of intermediary inputs adjusted for price changes. We find that it has improved significantly since 2002 in most sectors in core manufacturing, although not for all. The second factor that has offset increases in input prices and wages is labor productivity growth, measured here as value added per worker. The third factor has been that net taxes, as a share of gross output, have gone down by almost 1 percentage point, further contributing to the improved profit margin. We also analyze whether sectors that faced higher cost pressures have made larger efforts to offset the cost pressure coming in via input prices. Our results suggest that this has been the case, although we need to interpret the results carefully. Against a backdrop of concerns about different kinds of cost pressures impinging on China's industry, our analysis is useful in identifying the key source of cost pressures. Our results suggest that the dominant source has been via higher raw material prices. Since 2002, average weighted output prices have risen significantly less than average weighted input prices. The gap between output price rises and input price rises increased from 4 percentage points in 2005 to 4.3 percentage points in 2006 for core 2 manufacturing.1 That this was possible while at the same time increasing the average profit margin shows that the ability of China's industry to offset rising raw material prices by increasing efficiency has so far remained undiminished. II. Conceptual Framework and data We combine information on the cost structure from the input output (IO) tables with time series data on those cost items to analyze the changes in the cost structure. We get information on the revenue and cost structure of different sectors of China's industry using the input-output table of 2002.2 We identify time series data on prices, wages, employment, taxes, and depreciation that best proxies the concepts used in the input output tables. These time series come from the enterprise survey and statistics on employment, wages, taxes, and investment of the National Bureau of Statistics (NBS). We focus on industry, and within that, on manufacturing.3 The industry classification in our analysis is the same as the one used in input-output table. The data on total gross output in the IO table matches conceptually the data on sales revenue from the enterprise survey. The absolute levels are not the same, due to differences in coverage. But we assume that changes over time are sufficiently similar between the 2 data sets. Intermediate input in the IO tables is the value of all the intermediate goods and services consumed to produce output. Each sector consumes intermediate input from all other sectors, at sectoral prices. Value added is the newly added value in the production process. It can be broken down into compensation of employees, operating surplus, depreciation of capital, and net taxes on production. These items are assumed to develop over time in the same way as the following items from other NBS statistics: business profit, depreciation, labor cost (wages * employment), and various taxes and extra charges. Firm's profit Let i be sector i's profit, pi the output price (PPI), Yi output, rij the price of input from sector j, Mi the amount of input from sector j, wi the wage, Li employment, i depreciation, and ti intermediate input. i = piYi - rijMij - wi Li- di- ti (1) Dividing both sides by piYi gives margi = 1 - (rij p / i) * (Mij Yi) - (wi pi )* (Yi. Li ) - di p Yi) ­ ti p Yi) / / / / ( i / ( i (2) 1In this paper, core manufacturing refers to industry excluding mining and quarrying, coking, gas and petroleum refining, production and supply of electric power, heat power and water industries. 22002 is the most recent year for which we could find the IO table. 3Manufacturing firms include mining and quarrying, energy and utility enterprises. 3 Where margi is the profit margin. In Equation (2), rij / pi is the relative price of inputs from sector j into sector i, compared to output prices of sector i, Mij / Yi is the amount of input from sector j into sector i per unit of (real) output, wi / pi is the real wage for sector i, and Yi / Li is labor productivity in sector i. We analyze economic developments in China's industry for the period 1999-2006. Our industry classification follows the classification of the input-output table with 17 sectors/commodities. Table 1 shows how we mapped the raw material price data and the producer price data to this industry classification. Table 1: Mapping three sets of data Intermediate input cost Input-Output Table Output price Purchasing Price index of Producer Raw Materials Price Index Farm and Sideline Products Agriculture - Fuels and Power Mining and Quarrying Coal Farm and Sideline products Foodstuff Food Textile Materials Textile, Sewing, Leather and Furs Average of Textile, Products Tailoring, and Leather Timber and Paper Pulp Other Manufacturing 1/ Average of Timber, Paper, Cultural, Educational and Handcrafts Article Fuels and Power Production and Supply of Electric Power Power, Heat Power and Water Fuels and Power Coking, Gas and Petroleum Refining Petroleum Chemical Materials Chemical Industry Chemical Building Materials Building Materials and Non-metal Building Materials Mineral Products Average of Ferrous Metal Materials Metal Products Metallurgical and Non ferrous Metal & Electric Wire Ind Raw Materials & Semi Finished Machinery and Equipment Machine Building Pdts Consumer Price Index: Services Construction CPI: Services Transportation, Post and Telecommunication Services CPI: Services Wholesale and Retail Trades, Hotels and Catering Services CPI: Services Real Estate, Leasing and Business Services CPI: Services Banking and Insurance CPI: Services Other Services 1/ Other manufacturing include furniture, paper, and cultural, educational and handcrafts light manufacturing. 4 While we have tried to find the best mapping, we are aware that it is a rough estimate. In particular, the mapping is constrained by the fact that in some cases the raw material prices are at a rather aggregate level, lacking a more detailed breakdown. For instance, the mining and quarrying sector includes ferrous and non-ferrous metal ore mining, in addition to coal mining and petroleum and natural gas's extraction. Price developments of raw materials coming from these sectors differ substantially. The raw materials price index most closely associated with input costs from mining and quarrying is the one of fuels and power, as coal and petroleum make up 69 percent of the output of that sector (according to the 2002 IO table). But, the price of metal ore mining is not considered when we estimate input cost from the sector. This is a problem especially for periods when energy prices and those of metal ore have not moved in the same direction. For the period that we analyze, prices of energy and metal ore fortunately move broadly in the same direction. The input output tables provide information on the composition of output in different sectors in terms of intermediate input from other sectors and value added. As a share of gross output, enterprises in China's manufacturing industries use 70 percent of intermediate input from other sectors and create 30 percent of value-added (Table 2). Value added comprises employees' compensation (11 percent), profit (8 percent), taxes (6 percent) and depreciation (5 percent). The third column in table 2 reports inputs from different sectors as a share of total intermediate inputs from industry--that is, not including inputs from sectors such as construction, transportation etc. These weights we used to reproduce the overall raw material index. We use the input coefficients of China manufacturing industry from the 2002 input output table (see below) as a basis. The time series data, for the period 1999-2006, were estimated using information on the change in revenues, earnings, depreciation, tax and operating surplus taken from enterprise survey data, China labor statistical yearbook and China statistical yearbook. The details are follows: Revenue = Total Output 2002 * (1+growth rate of revenue series from the Enterprise Survey) Profit = Operating Surplus 2002 * (1+growth rate of profit series from the Enterprise Survey) Earnings = Compensation of Employees in 2002 * (1+growth rate of earnings series from the China Labour Statistical Yearbook ) Depreciation = Depreciation of Fixed Assets in 2002 * (1+growth rate of our estimate for depreciation). This estimate is urban fixed asset investment from the China Statistical Yearbook minus increase in net fixed asset of Enterprise Survey ) Tax =Net Taxes on Production in 2002 *(1+growth rate of sales taxes and VAT from the Enterprise Survey) 5 Table 2: Summary of Input-Output coefficients for the Manufacturing Industry As share As share of As share of of gross intermediate intermediate output output input from industry 100 Intermediate Input 70 100 Agriculture 5 7 8 Mining and Quarrying 6 8 10 Foodstuff 1 2 3 Textile, Sewing, Leather and Furs Products 4 6 7 Other Manufacturing 4 5 7 Production and Supply of Electric Power, Heat Power and Water 3 4 5 Coking, Gas and Petroleum Refining 2 2 3 Chemical Industry 10 14 17 Building Materials and Non-metal Mineral Products 1 2 2 Metal Products 10 14 17 Machinery and Equipment 12 18 22 Construction 0 0 Transportation, Post and Telecommunication Services 3 5 Wholesale and Retail Trades, Hotels and Catering Services 5 7 Real Estate, Leasing and Business Services 2 3 Banking and Insurance 1 2 Other Services 1 1 Total Value-added 30 Compensation of Employees 11 Net Taxes on Production 6 Depreciation of Fixed Assets 5 Operating Surplus 8 Source: NBS, Input-output table (2002) The nominal intermediate input is calculated as a residual, as total output minus all other costs and profits. The machinery and equipment industry is the most important intermediate input into other sectors, making up 18 percent of the total, with the chemical and metal industry both contributing 14 percent. Therefore, output price increases of these industries would most affect cost pressures from intermediate inputs. Machinery and equipment prices have increased little since 2002, which has dampened cost pressures. III. Price increases International commodity prices have risen strongly since 2002, particularly since 2003. Fuelled by high oil prices, prices of energy rose by an average of 29 percent per year in 2004-2006, in RMB terms (Table 3). Non-energy commodity prices also rose 6 substantially--some 17 percent on average in 2004-2006--, with prices of metals and minerals rising a particularly rapid 36 percent per year on average in 2004-2006. Table 3: International commodity prices and China raw material prices (RMB) (Annual percent change) 2002 2003 2004 2005 2006 2004-6 International commodity prices Energy 2.4 15.9 30.6 40.1 17.1 28.9 Non-energy 5.3 10.2 17.4 12.4 21.2 16.9 Metals and Minerals -3.1 12.6 37.1 25.5 46.8 36.2 Purchasing Price Index of raw materials Overall -2.3 4.8 11.4 8.3 6.0 8.5 Fuels and power 0.2 7.4 9.7 15.0 11.8 12.2 Non-ferrous Metals4 -3.5 5.3 20.1 14.0 30.8 21.5 Source: World Bank, National bureau of Statistics Domestic raw material prices in China also rose significantly, although they lagged international prices. The overall purchasing price index of raw materials rose 8.5 percent per year on average in 2004-2006, with the sub-index "fuel and power" rising 12 percent per year and that of "non-ferrous metals" 21.5 percent. There are 2 key reasons why overall raw material prices in China have not risen as much as the international commodity price indices. First, looking at prices of individual products, while some domestic prices have followed closely international prices, other have not (Table 4). Prices of copper, nickel, rubber, and, to some extent, zinc and coal have followed world market prices closely. However, prices of aluminum, steel, and, importantly, oil and oil products, have risen significantly less than international ones. For oil and oil products, the reason behind the difference is that the government controls the price of refined products administratively. For others, such as steel and aluminum, domestic market conditions may have affected price movements over and beyond the impact of international prices. Comparing the absolute price level of commodities is difficult, because of the difficulty of finding the exact same product. However, prices of some commodities are quite similar to international prices, including copper and, to some extent zinc and aluminum (Table 5). Rubber even appears to be more expensive in China. The second reason is that the composition of China's raw material price index differs from the international commodity price indices. In part this reflects different consumption patterns and economic structures. Another reason is that China's "raw material prices" include prices of intermediary goods such as chemicals, metals, and "machinery and equipment".5 Indeed, we are able to reproduce the overall raw material price time series by using the weights from the 2002 input output table from Table 2 (Figure 1). And, 4Non-ferrous metals include aluminum, copper, gold, nickel, silver, tin, and zinc 5China's raw material price index is estimated based on goods used by industrial enterprises and captures over 900 products in 9 categories. The index measures what happens to prices paid by industrial enterprises when they purchase inputs such as raw materials, fuels, and power. 7 indeed, prices of intermediary goods such as chemicals and "machinery and equipment" prices have not increased as strongly as primary commodity prices. Table 4: World Commodity Prices vs. China Wholesale prices (Annual percent change) 2002 2003 2004 2005 2006 2004-6 Aluminum World -6.4 6.0 19.9 9.6 31.7 20.0 China -4.3 1.8 10.5 2.0 22.6 11.4 Copper World -1.1 14.1 61.1 27.1 77.8 53.8 China -6.1 8.3 46.2 29.1 79.7 50.2 Nickel World 13.9 42.2 43.6 5.6 60.1 34.4 China -9.8 14.3 52.0 11.9 47.3 35.8 Zinc World -12.1 6.3 26.6 30.5 130.7 56.2 China -8.9 -4.3 25.9 19.9 102.4 45.1 Steel World 2.9 8.3 44.2 13.5 -4.0 16.3 Round steel China -3.7 21.8 27.9 -5.4 -7.7 3.7 Rubber Singapore 33.0 41.6 20.4 14.0 36.6 23.3 China -2.2 20.8 28.7 6.0 48.2 26.5 Logs Malaysia 2.7 14.6 5.4 1.9 14.7 7.2 China -2.6 -0.8 3.9 6.8 -1.4 3.0 Coal Australia1 -21.7 3.2 102.7 -10.9 0.4 21.9 China 6.3 8.3 25.8 21.3 7.3 17.9 Crude Oil World 2.1 16.1 30.5 40.2 17.2 28.9 Diesel Oil China- -1.2 11.5 11.6 15.3 17.5 14.8 Natural gas U.S. -15.0 61.8 7.3 49.4 -26.7 5.5 China -4.5 10.3 1.5 0.3 13.6 5.0 Source: World Bank, NBS Figure 1: The official PPI and a proxy using our price and weight estimates (year on year change) 115.0 raw material prices intermediate input weight *raw material prices 110.0 105.0 100.0 95.0 90.0 85.0 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source: NBS, authors' estimates 8 Table 5: Price levels of raw materials on international markets and in China (2006) Commodity Region Unit In RMB Aluminum World Per ton 20491 China Per ton 19828 Copper World Per ton 53599 China Per ton 54452 Zinc World Per ton 26113 China Per ton 23234 Wire Steel World Per ton 4262 China Per ton 3195 Rubber Singapore Per ton 16818 China Per ton 20421 Logs Malaysia Cum 1909 China Per cube m 930 Coal Australia1/ Per ton 391 China Per ton 558 Crude Oil World Per ton 4102 Diesel Oil China- Per ton 5043 Natural gas U.S. Mmbtu 53 China Cube m 2 1/ Commodities in bold are closer to each other in terms of definition and unit. Nonetheless, even measured by China's raw material price index, input prices in manufacturing have risen substantially. We constructed time series for the intermediate input costs of all sectors of industry, using the raw material price data and the intermediate input value weights of the 2002 input output table (Table 2). The weighted input prices are shown in Figure 2. The overall numbers are affected significantly by the sectors mining and quarrying and petroleum refining. For analytical purposes we define an aggregate "core manufacturing". This excludes mining and quarrying, electric power, and coking, gas and petroleum refining. Thus, sectors that are heavy users of the raw materials that have risen most in prices--for instance, non ferrous metals and fuel and power--have faced the highest overall increases in intermediate input costs. Overall, input costs have increased most for the metal industry, petroleum refining, and utilities.6 Output prices (PPI, or factory gate) have generally also increased, although much more for some than for others (Figure 3). The impact of input price increases on sectors depends on how much output price rises offset the input price increases. The relative prices are shown in Figure 4 and Table 6. Thus, with the factory gate prices for the metal sector also increasing substantially, the relative ("real") cost of inputs for the metal sector did hardly increase (it was 3 percent higher in 2006 than in 2002). 6For instance, machinery and equipment consumed metals at 19 percent of the value of total intermediate inputs. 9 Figure 2: Weighted input prices (raw materials) 2002 = 100 1/ 150 Manufacturing 150 Core manufacturing Mining and quarrying Textile 140 Utility Chemical 140 Building materials Petroleum refining Metal products Foodstuff 130 Machinery and equipment Other 130 120 120 110 110 100 100 90 90 2003 2004 2005 2006 2003 2004 2005 2006 Source: NBS, authors' estimates. 1/ In this and all other figures, the aggregate "manufacturing" refers to mining, manufacturing and utilities. Figure 3: Output prices 2002 = 100 200 Manufacturing 200 Core manufacturing Mining and quarrying Textile 180 Utility 180 Chemical Petroleum refining Building materials Metal products Foodstuff 160 Machinery and equipment Other 160 140 140 120 120 100 100 80 80 2003 2004 2005 2006 2003 2004 2005 2006 Source: NBS, authors' estimates. On the other hand, the relative price of input costs for the machinery and equipment sector worsened significantly, because it consumes a lot of metal products (19 percent of the total), while its own output prices have been stuck at the 2002 level or worse over the four years as shown in Figure 4. Outside of core manufacturing, sectors such as petroleum refinery and mining and quarrying saw their factory gate prices rise more than the average price of input costs. 10 Table 6: Intermediate Input Cost 2002=100 Weighted input prices (ri) Relative input price to output price (ri p ) / i 2003 2004 2005 2006 2003 2004 2005 2006 Mining, manuf., and utilities 103.3 113.0 121.4 130.3 101.0 104.1 106.6 111.1 Core manufacturing 103.1 112.4 119.2 126.0 102.4 107.2 111.6 116.3 Mining and Quarrying 103.3 111.1 120.1 128.1 96.5 89.6 81.9 75.3 Utilities7 104.1 111.9 123.8 134.9 103.2 108.3 115.0 121.9 Petroleum Refining 106.1 115.6 131.2 145.5 91.8 87.5 81.2 75.2 Foodstuff 105.2 117.5 120.8 126.1 104.0 109.2 111.2 115.5 Textile 102.4 109.4 113.5 117.5 101.8 106.3 109.1 111.2 Other 102.1 109.1 115.0 120.6 103.3 109.2 113.8 118.1 Chemical 103.4 112.2 121.7 127.9 101.1 101.8 103.4 110.0 Building Materials 103.1 111.0 120.8 130.2 103.5 107.7 116.4 123.4 Metal 105.0 119.3 131.7 147.6 98.3 95.6 98.8 103.1 Machinery and Equipment 102.0 111.2 118.0 126.2 105.1 115.3 123.3 131.8 Source: NBS, authors' estimates. Figure 4: Relative prices (input/output) 2002 = 100 150 150 Manufacturing Core manufacturing 140 Mining and quarrying 140 Textile Utility Chemical 130 Petroleum refining 130 Building materials Foodstuff Metal products 120 Other 120 Machinery and equipment 110 110 100 100 90 90 80 80 70 70 2003 2004 2005 2006 2003 2004 2005 2006 Source: NBS, authors' estimates. IV. Wage increases Wage growth has also been fast. The official sectoral wage data indicates that wages have risen by 10-15 percent per year since 2002 in most sectors, with wage growth in the booming mining and quarrying sector particularly high (Figure 5). Comparing nominal wage developments with output prices, real wage increases have also been rapid in almost all sectors. The official wage data only covers formal employment at SOEs and some large private sector enterprises. Including other forms of employment would reduce the average wage increase, since informal employees and employees outside of SOEs and 7Indicate production and supply of electric power, heat power and water. 11 large firms have seen lower wage increases than the formally employed. At the same time, including these forms of employment--which grew faster than formal employment-- would increase the average increase in employment, which offsets the effect on wages. We make the rough assumption that these two effects cancel each other out. Figure 5: Nominal wages (percent change) 30 Manufacturing 30 28 Mining and quarrying 28 Core manufacturing Utility Textile 26 Petroleum refining 26 Chemical 24 Foodstuff 24 Metal products Other 22 22 Machinery and equipment Other 20 20 18 18 16 16 14 14 12 12 10 10 8 8 2002 2003 2004 2005 2002 2003 2004 2005 Source: NBS, authors' estimates. V. Impact of cost pressures on industry and response Despite rapid increases in prices of intermediate inputs and wages, profit margins have actually continue to rise in many sectors since 2002 (Figure 6). Combined with rapid growth of turnover/sales in industry, this has resulted in impressive profit growth (Table 7). Figure 6: Profit margin (percent) 13 Manufacturing 40 8 Core manufacturing Utility Textile 11 Petroleum Refining 35 7 Chemical Foodstuff Building Materials 9 Other mining and quarrying (right axis) 30 6 Metal products Machinery and Equipment 7 25 5 5 20 4 3 15 3 1 2 -1 10 1 -3 5 1999 2000 2001 2002 2003 2004 2005 2006 1999 2000 2001 2002 2003 2004 2005 2006 Source: NBS, authors' estimates. 12 Table 7: China's manufacturing industry (percent change) 2002 2003 2004 2005 2006 2002-6 Sales 16 38 34 30 26 29 Profit 22 50 40 22 30 33 Source: NBS. What has happened to offset the cost pressures? First, the technical efficiency of the usage of intermediate inputs real terms has improved substantially over time. We estimate what has happened to the technical efficiency of using intermediate inputs over time by putting together the time series data on input prices, output prices, wages, employment, and profit margins. We calculate time series for all industrial sectors of the amount of intermediate inputs per unit of output in constant prices. Referring to equation (2), we solve it for Mij Yi in prices of 2002. We / interpret a decline in this indicator as an improvement in the technical efficiency of using intermediate inputs. This improved technical efficiency and upgrading of the production structure leads to more profit per unit of sales. Our estimation suggests that since 2002, when the raw material price increases accelerated, technical efficiency gains have been large in core manufacturing (Table 8). The implied intermediate input share of total output in constant prices declined on average from almost 74 percent in 2003 to 67 in 2006, with a significant amount of variation between sectors. That is, most of manufacturing industry improved substantially the technical efficiency of the usage of real intermediate inputs. Interestingly, this was not the case for mining and quarrying, petroleum refining, and metals. Second, labor productivity increased rapidly in core manufacturing industries during this period. Table 9 shows labor productivity growth across sectors in industry, calculated as value added in real terms divided by employment. The furniture, paper and handcraft industries achieved the most rapid productivity increase. As a result, the labor cost burden, estimated as the ratio of wage costs to total output, decreased significantly over the period 2003-2006 (Table 10). 13 Table 8: The ratio of intermediate input to output (percent) 2002 2003 2004 2005 2006 Change1/ Mining, manuf., and utilities (Nominal intermediate input ratio) 70.1 73.2 73.1 74.7 75.1 4.9 (Real intermediate input ratio) 70.1 72.5 70.2 70.1 67.5 -2.6 Core manufacturing 72.9 75.7 76.2 78.1 78.0 5.1 72.9 73.9 71.1 70.0 67.1 -5.8 Mining and Quarrying 42.2 41.4 40.5 39.6 42.0 -0.2 42.2 42.9 45.2 48.3 55.8 13.6 Utilities 49.9 56.8 52.5 52.5 51.1 1.2 49.9 55.0 48.4 45.7 41.9 -8.0 Petroleum Refining 82.6 81.7 81.0 94.9 98.3 15.7 82.6 88.9 92.5 116.9 130.7 48.1 Foodstuff 68.9 86.6 81.5 81.7 80.9 11.9 68.9 83.3 74.7 73.5 70.0 1.1 Textile 75.3 75.7 75.3 73.9 73.4 -1.9 75.3 74.4 70.8 67.7 66.0 -9.3 Other 65.0 70.4 67.8 70.7 67.1 2.2 65.0 68.2 62.1 62.1 56.8 -8.1 Chemical 73.1 71.6 72.5 78.5 80.0 6.9 73.1 70.8 71.2 75.9 72.7 -0.4 Building Materials 67.1 68.4 72.0 74.4 73.3 6.2 67.1 66.0 66.9 64.0 59.4 -7.7 Metal 75.8 76.2 79.7 82.0 82.8 7.0 75.8 77.6 83.4 83.0 80.3 4.5 Machinery and Equipment 75.1 77.3 78.8 79.1 78.5 3.4 75.1 73.5 68.3 64.1 59.5 -15.6 Source: NBS, authors' calculation. 1/ from 2002 to 2006 Table 9: The growth rate of labor productivity across industry (Percent) 2002 2003 2004 2005 2006 2003-6 Mining, manuf., and utilities 26 24 27 14 16 20.2 Core manufacturing 29 29 24 14 19 21.4 Mining and Quarrying -1 21 37 21 5 20.5 Utilities 9 9 33 23 21 21.4 Petroleum Refining 77 20 35 -69 -67 -36.5 Foodstuff 28 -5 64 25 21 24.0 Textile 19 23 24 25 13 21.2 Other 40 39 37 19 32 31.8 Chemical 35 36 20 -2 15 16.6 Building Materials 24 28 15 14 28 21.1 Metal 24 38 14 12 13 18.8 Machinery and Equipment 33 32 18 20 24 23.2 Source: China Labour Statistical Yearbook, author's calculation. 14 Table 10: The ratio of labor cost to total gross output in current prices (Percent) 2002 2003 2004 2005 2006 Change Mining, manuf., and utilities 11.5 9.5 8.3 7.5 7.1 -4.3 Core manufacturing 10.7 8.8 7.6 6.9 6.3 -4.5 Mining and Quarrying 24.9 22.0 19.7 16.9 15.0 -9.9 Utilities 15.6 10.5 9.7 8.6 8.2 -3.4 Petroleum Refining 5.3 4.5 3.5 3.1 2.8 -2.5 Foodstuff 8.3 4.3 3.8 3.5 3.3 -5.0 Textile 11.8 10.7 10.3 10.0 9.4 -2.4 Other 12.8 9.4 8.6 7.5 6.9 -5.9 Chemical 9.5 8.4 7.0 6.0 5.6 -3.9 Building Materials 17.0 15.1 12.9 11.6 10.5 -6.6 Metal 10.7 8.6 6.4 5.7 4.9 -5.7 Machinery and Equipment 10.3 8.9 7.8 7.3 6.6 -3.8 Source: China Labour Statistical Yearbook, author's calculation. Playing a smaller role, a reduction in the impact of taxation also help boost profitability. We did not find a significant change over time in the impact of depreciation on profitability. Table 11. Input cost pressures and technical efficiency of using intermediate inputs in industry (2002-06, annual average change) Weighted Efficiency Wage Labor relative input of using (nominal) productivity price (%) 1/ intermediary (%) (nominal) inputs (%p) 2/ (%) Manufacturing 2.7 0.7 14.0 21.3 Core manufacturing 3.8 1.5 12.7 22.9 Mining and Quarrying -6.8 -3.2 21.0 16.0 Utility 5.1 2.1 14.4 18.9 Petroleum Refining -6.9 -10.3 12.6 -22.1 Foodstuff 3.7 -0.3 11.6 24.9 Textile 2.7 2.4 12.0 20.7 Other 4.2 2.1 13.6 33.4 Chemical 2.4 0.1 12.3 20.0 Building Materials 5.4 2.0 11.9 21.7 Metal products 0.8 -1.1 14.1 19.8 Machinery and Equipm. 7.1 4.2 12.7 25.2 Source: NBS, authors' estimates. 1/ For each sector, the weighted raw material price increase is estimated using weights from the 2002 input output table. 2/ Estimated improvement technical efficiency of using intermediary inputs. 15 A look at the patterns across sectors suggests that the effort to increase the technical efficiency of using intermediary inputs, improve labor productivity, and limit wage increases have been the largest in those sectors with the highest cost pressures. This is illustrated in Figures 7-9. Figure 7 suggests that sectors with a higher increase in relative input prices have paid out lower wage increases. Figure 8 suggests that sectors with higher increases in relative input prices have seen higher labor productivity growth. Figure 9 suggests that sectors with higher relative input prices have seen a large reduction in the usage of intermediary inputs per unit of output. For instance, sectors with particularly high pressure from input price rises, such as building materials and machinery and equipment industries, have limited wage growth, boosted labor productivity, and improved the technical efficiency of the production process more than other sectors. Figure7: wage growth is lower where cost Figure 8: labor productivity growth pressures are higher higher where cost pressures higher 25 40 30 20 20 15 ytivit 10 eg uc Wa odrPr y = 2.1194x + 15.171 0 10 boaL R2 = 0.3722 -10 y = -0.4493x + 14.196 5 R2 = 0.4836 -20 0 -30 -10 -8 -6 -4 -2 0 2 4 6 8 -10 -8 -6 -4 -2 0 2 4 6 8 Relative input prices Relative input prices Source: authors' estimates. V. Conclusion The main contribution of this paper has been to provide a general framework for analyzing how improvements in the cost structure and efficiency have partly or fully offset the impact on profit of a rapid rise in raw material prices and wages in China's industry. Caveats are in place. Our findings are qualified by the fact that prices and input and output figures are broad and that 2002 was the most recent IO table available. There are other data concerns. For instance, the data on wages and employment has limited coverage. Nonetheless, we think that the data contains useful information on changes over time. Our analysis helps in understanding the impact of cost increases and how they have been partly or fully offset, as well as the impact across industries. 16 Figure 9. Improvement in technical efficiency is higher where cost pressures are higher 60 50 40 nputI y = -3.3783x + 6.0846 teaide R2 = 0.6218 30 mr tenItneic 20 10 ffieoClaeR 0 -10 -20 -10 -8 -6 -4 -2 0 2 4 6 8 Relative input prices Source: NBS, authors' estimates. China's industrial sector has experienced hefty cost pressures in recent years coming from rapid increases in raw material prices and wages. At the same time, profit margins have largely continued their trend rise. China's industrial enterprises have been able to cope with the cost pressures because of rapid growth of productivity and efficiency and output price increases. We quantify the productivity and efficiency gains. Our results suggest that those sectors with the largest cost pressures appear to have made the largest effort in offsetting the cost pressures. Against a backdrop of concerns about different kinds of cost pressures impinging on China's industry, our results are also useful in identifying the key source of cost pressures. Our results suggest that the dominant source of cost pressures is via higher raw material prices. Since 2002, average weighted output prices have risen significantly less than average weighted input prices. The gap between output price rises and input price rises increased from 4 percentage points in 2005 to 4.3 percentage points in 2006 for core manufacturing (Table 6), while the average profit margin rose in that year. This shows that the ability of China's industry to offset rising raw material prices by increasing efficiency has so far remained undiminished. 17