• 検索結果がありません。

Results and Findings

ドキュメント内 Juita Binti Mohamad (ページ 190-200)

Chapter 7: Employment Inequality and Skilled - biased Technological Change: A Plant Level

7.6 Results and Findings

Table 7.2: Employment Equations of Production Workers Dependant Variable: log (total production workers)

Source: Author`s calculation based on data obtained from the Annual Manufacturing Survey sample data Looking at the results above in Table 7.2 let us analyze the findings step by step. In Column 1, the specification in this column includes only plant level variables without including the additional plant-level controls such as the FDI dummy, large plantsize dummy, holding type

Dep. Variable = ln(total production

worker)

1 2 3

Logreal wage productionworker -0.16*** -0.14*** -0.13***

(0.02) (0.01) (0.01)

Logreal Machinery& Equipment 0.11*** 0.09*** 0.10***

(0.004) (0.004) (0.004)

Logreal sales 0.51*** 0.43*** 0.47***

(0.005) (0.004) (0.006)

Exportsdummy 0.241*** 0.18*** 0.11***

(0.02) (0.02) (0.02)

R&Ddummy -0.03*** -0.03 -0.05***

(0.05) (0.02) (0.02)

Patentdummy 0.05 0.01 -0.01

(0.02) (0.02) (0.04)

FDIdummy -0.005 -0.01***

(0.02) (0.02)

LargeFirm dummy 0.86*** 0.70***

(0.02) (0.02)

Privatefirmdummy 0.13*** 0.14***

(0.015) (0.015)

State Group1 dummy -0.101*** -0.086***

(0.012) (0.012)

Observations 21745 21745 21745

R-squared 0.74 0.76 0.79

From the table above we can see that here the relationship between R&D activities and log of total production workers employed is negative and significant at 1% level. It is observed that there is a positive relationship between the usage of patents and log of total production workers employed. However, the coefficient is insignificant which implies that usage of patents does not influence the total number of production workers employed in plants in Malaysia.

Patents dummy does not have a significant impact on influence the total number of production workers employed for all the regression specifications, which can be seen in all of the 3 columns.

We can see that the relationship between the log real wage of production workers and total production workers employed is negative. This is consistent with our expectations whereby the negative values of wages are in line with the standard requirement for the relationship between wages and labor demand. As the price of the factor increases, and in this case the production workers, the less the demand is for that factor. From the table above we find that a 1% increase in log real wages of production workers will lead to a decrease of demand by 0.16%

as reported in the first column. The coefficient is negative and significant at 1% level.

The remaining explanatory variables all show to have a statistically significant employment enhancing effect. We also find that here, the relationship between log real machinery and equipment and log total production workers employed is positive and significant at 1% level. We find here that machinery and equipment has a positive impact on log total production workers employed. A 1% increase in the size of machinery and equipment owned in stocks and assets by a plant will lead to a 0.11% increase in the total production workers employed. This indicates that the higher the intensity of machinery and equipment usage in a plant, the higher the demand for production workers.

In column 2 the specifications include the main plant level variables and the additional plant-level controls such as the FDI dummy, large plantsize dummy, private plant dummy and state dummy. Year fixed affect is also included in this specification for a more robust finding.

From the table above we can see that here the relationship between existence of R&D activities and log of total production workers employed is negative but not significant. It is observed that there is a positive relationship between the usage of patents and log of total

production workers employed. However the coefficient is insignificant which implies that usage of patents do not influence the total number of production workers employed in plants in Malaysia. Patents dummy do not have a significant impact on influence the total number of production workers employed for all the regression specifications, which can be seen in all of the 3 columns. From the table above we can see that here the relationship between existence of FDI and log of total production workers employed is positive but not significant. We find here that existence of FDI have a positive impact on log of total production workers employed. If a plant is an FDI, this will lead to a 0.005% decrease in the demand for production workers when compared to non-FDI plants. The FDI dummy coefficient is positive but is insignificant in affecting demand of total production.

We can see that the relationship between the log real wage of production workers and total production workers employed is negative. This is also consistent with our expectations whereby the negative coefficients of wages are in line with the standard requirement for the relationship between wages and labor demand. As the price of the factor increases, and in this the production workers, the less the demand is for that factor. From the table above we find that a 1% increase in log real wages of production workers will lead to a decrease of demand by 0.14%, which is slightly smaller than the coefficient reported in the first column. The coefficient is negative and significant at 1% level. The remaining explanatory variables all show to have a statistically significant employment-enhancing effect. We also find that here, the relationship between log real machinery and equipment and log total production workers employed is positive and significant at 1% level. We find here that machinery and equipment has a positive impact on log total production workers employed. A 1% increase in the size of machinery and equipment owned in stocks and assets by a plant will lead to a 0.09% increase in the total production workers employed.

We also find that there positive relationship between the size of machinery and equipment with total number of production workers employed. We can see that a 1% increase in the size of machinery and equipment owned in stocks and assets by a plant will lead to a 0.11% increase in the total production workers employed. This indicates that the higher the intensity of machinery

In column 3 the specifications include the main plant level variables and the additional plant-level controls such as the FDI dummy, large plantsize dummy, holding type dummy and state dummy. I have also included year 2000 dummy and industry fixed effects for a more robust finding.

From the table above we can see that here the relationship between existence of R&D activities and log of total production workers employed is positive and significant at 1% level.

We find here that existence of R&D activities have a positive impact on log of total production workers employed. If a plant is involved in exporting activities, this will lead to a 0.05% increase in the demand for production workers when compared to other plants not involved in any R&D activities. The R&D dummy coefficient is positive and significant at 1% level.

It is observed that there is a positive relationship between the usage of patents and log of total production workers employed. However the coefficient is insignificant which implies that usage of patents do not influence the total number of production workers employed in plants in Malaysia. Patents dummy do not have a significant impact on influence the total number of production workers employed for all the regression specifications, which can be seen in all of the 3 columns.

From the table above we can see that here the relationship between existence of FDI and log of total production workers employed is negative but significant. We find here that existence of FDI have a negative impact on demand of total production workers. If a plant is an FDI, this will lead to a 0.01% decrease in the demand for production workers when compared to non-FDI plants. FDI plants are not significant in affecting demand of total production.

We can see that the relationship between the log real wage of production workers and total production workers employed continues to be negative. As mentioned before this is consistent with our expectations whereby the negative values of wages are in line with the standard requirement for the relationship between wages and labor demand. As the price of the factor increases, and in this the production workers, the less the demand is for that factor. From the Table above we find that a 1% increase in log real wages of production workers will lead to a

decrease of demand by 0.13% as reported in the first column. The coefficient is negative and significant at 1% level but slightly smaller than the coefficients in columns 1 and 2.

We also find that here, the relationship between log real machinery and equipment and log total production workers employed is positive and significant at 1% level. We find here that machinery and equipment has a positive impact on log total production workers employed. A 1%

increase in the size of machinery and equipment owned in stocks and assets by a plant will lead to a 0.1% increase in the total production workers employed. This indicates that the higher the intensity of machinery and equipment usage in a plant, the higher the demand for production workers.

Here the relationship between log real sales and total production workers employed is positive and significant at 1% level. We find here real sales have a positive impact on log total production workers employed. A 1% increase in the size of real sales of a plant will lead to a 0.18% increase in the total production workers employed. This indicates that expansion of production and therefore sales, requires higher demand for production workers. The coefficient is slightly lower than in column 1. This indicates that other additional variables might have an effect on total production workers employed.

We can see that here the relationship between log real sales and total production workers employed is positive and significant at 1% level. We find here that machinery and equipment has a positive impact on log total production workers employed. A 1% increase in the size of machinery and equipment owned in stocks and assets by a plant will lead to a 0.47% increase in the total production workers employed.

We also find that here the relationship between exports dummy and log of total production workers employed is positive and significant at 1% level. We find here that exports have a positive impact on log of total production workers employed. If a plant is involved in exporting activities, this will lead to a 0.11% increase in the demand for production workers when compared to other types of plants such as mixed ownership plants and Malaysian-owned plants. The export dummy coefficient is positive and significant at 1% level.

It is observed that there is a positive relationship between the large-sized plant dummy and log of total production workers employed. This is consistent with our expectations whereby the positive values of large plant dummy are in line with the standard requirement for the relationship between size and labor demand. Here we see that the coefficient is significant at 1%

level. Large plants this will increase the demand of production workers by 0.7%. The coefficient remains significant in column 3, which implies that the size of the plant influences the total number of production workers employed in plants.

It is observed that there is a positive relationship between the private plant dummy and log of total production workers employed. Here we see that the coefficient is significant at 1%

level. Private plants will increase the demand of production workers by 0.14% if compared to other non-privately owned plants. This coefficient is slightly higher than the one in the previous column. The coefficient continues to be significant in columns 2 and 3, which imply that the holding type of plants influences the total number of production workers employed in plants in Malaysia.

From the table above we can see that here the relationship between plants located in State 1 dummy and log of total production workers employed is negative and significant at 1% level.

We find here if plants or plants are located in State Group 1, this has a negative impact on log of total production workers employed. If a plant is situated in either of the State Group 1, this will lead to a 0.09% decrease in the demand for production workers when compared to plants located in other group states. This implies that demand for production workers in plants and plants in State Group 1 are lower by 0.09% compared with plants in other Group States.

Table 7.3: Employment Equations of Non-Production Workers Dependant variable: log (total non-production workers)

Source: Author`s calculation based on data obtained from the Annual Manufacturing Survey sample data

Dep.Variable = ln(total

non-productionworker) 1 2 3

Logreal wage non-productionworker 0.10*** 0.09*** 0.078***

(0.01) (0.01) (0.01)

Logreal Machinery& Equipment 0.14 *** 0.11*** 0.11***

(0.003) (0.003) (0.003)

Logreal sales 0.43*** 0.33*** 0.34***

(0.004) (0.004) (0.005)

Exportsdummy 0.23*** 0.12*** 0.09***

(0.014) (0.013) (0.013)

R&Ddummy 0.03*** 0.23*** 0.20***

(0.04) (0.02) (0.02)

Patentdummy 0.33*** 0.05 0.04

(0.017) (0.04) (0.03)

FDIdummy 0.20*** 0.13***

(0.02) (0.02)

LargeFirm dummy 0.97*** 0.86***

(0.02) (0.02)

Privatefirmdummy 0.399*** 0.35***

(0.01) (0.01)

State Group1 dummy 0.05*** 0.03***

(0.0095) (0.0096)

Observations 22091 22091 22091

R-squared 0.81 0.84 0.85

Looking at the results above in Table 7.3 let us analyze the findings step by step. In Column 1, the specification in this column includes only plant level variables without including the additional plant-level controls such as the FDI dummy, large plantsize dummy, holding type dummy and state dummy.

From the table above we can see that here the relationship between existence of R&D activities and log of total production workers employed is positive and significant at 1% level.

We find here that existence of R&D activities have a positive impact on log of total non-production workers employed. If a plant is involved in exporting activities, this will lead to a 0.03% increase in the demand for non-production workers when compared to plants not involved in any R&D activities. The R&D dummy coefficient is positive and significant in affecting demand of total production.

It is observed that there is a positive relationship between the usage of patents and log of total non-production workers employed. The coefficient is significant which implies that usage of patents does influence the total number of production workers employed in plants in Malaysia.

This result is not robust as the Patents dummy significance dies out as we move along the regression specification toward the right side of the table. This implies that the patents do not have a significant impact on influencing the total number of non-production workers employed in specifications 2 and 3 in the table above.

We can see that the relationship between the log real wage of non-production workers and total non-production workers employed is positive. This is however not consistent with our expectations whereby the negative values of wages are in line with the standard requirement for the relationship between wages and labor demand. In the case of non-production worker as the price of the factor increases, and in this case real wages, the higher the demand is for that factor.

This type of relationship is observed with luxury products, whereby the higher the prices are for these types of goods, the higher the demand it is for these products. This implies that non-production workers or skilled workers are seen as a “luxury factor” in Malaysia`s manufacturing sector. Skilled workers are relatively scarce in a developing country such as Malaysia and an increase in real wages of 1% would lead to a 0.1% increase in demand for skilled workers.

In column 2 the specifications include the main plant level variables and the additional plant-level controls such as the FDI dummy, large plantsize dummy, holding type dummy and state dummy. Year fixed affect is also included in this specification for a more robust finding.

From the table above we can see that here the relationship between existence of R&D activities and log of total non-production workers employed is positive and significant. We find here that existence of R&D activities have a positive impact on log of total non-production workers employed. If a plant is involved in exporting activities, this will lead to a 0.23% increase in the demand for non-production workers when compared to plants not involved in any R&D activities. The R&D dummy coefficient is positive but is insignificant in affecting demand of total non-production workers.

It is observed that there is a positive relationship between the usage of patents and log of total production workers employed. However the coefficient is insignificant, which implies that usage of patents do not influence the total number of non- production workers employed in plants in Malaysia. Patents dummy does not have a significant impact on influencing the total number of non-production workers employed for all the remaining regression specifications in columns 2 and 3.

From the table above we can see that here the relationship between existence of FDI and log of total non-production workers employed is positive and highly significant. We find here that existence of FDI has a positive impact on log of total non-production workers employed. If a plant is an FDI, this will lead to a 0.20% increase in the demand for production workers when compared to non-FDI plants. The FDI dummy coefficient is positive and very significant at 1%

level, in affecting demand of total production.

We can see that the relationship between the log real wage of non-production workers and total non-production workers employed is positive. This is however not consistent with our expectations whereby the negative values of wages are in line with the standard requirement for the relationship between wages and labor demand. In the case of non-production worker as the price of the factor increases, and in this case real wages, the higher the demand is for that factor.

As mentioned before this type of relationship is observed with luxury products, whereby the

higher the prices are for these types of goods, the higher the demand it is for these products. This implies that non-production workers or skilled workers are seen as a “luxury factor” in Malaysia`s manufacturing sector. Skilled workers are relatively scarce in a developing country such as Malaysia and an increase in real wages of 1% would lead to a 0.09% increase in demand for skilled workers.

It is observed that there is a positive relationship between private plants and plants dummy and log of total non-production workers employed. Here we see that the coefficient is significant at 1% level. We find that Private plants increase the demand of non-production workers more compared to other non-privately owned plants. The coefficient continues to be significant in columns 2 and 3, which imply that private plants and plants influences the total number of non- production workers employed in plants in Malaysia.

From the table above we can see that here the relationship between plants located in State 1 dummy and log of total non-production workers employed is positive and significant at 1%

level. We find here if plants or plants are located in State Group 1, this has a positive impact on log of total production workers employed. If a plant is situated in the State Group 1, this will lead to a 0.4% increase in the demand for non-production workers when compared to plants located in other group states. This implies that demand for non-production workers in plants and plants in State Group 1 is higher by 0.4% compared with plants in other Group States.

In column 3 the specifications include the main plant level variables and the additional plant-level controls such as the FDI dummy, large plantsize dummy, holding type dummy and state dummy. I have also included year fixed affect and industry fixed effects for a more robust finding.

From the table above we can see that here the relationship between existence of R&D activities and log of total non-production workers employed is positive and significant. We find here that existence of R&D activities have a positive impact on log of total non-production workers employed. If a plant is involved in exporting activities, this will lead to a 0.23% increase in the demand for non-production workers when compared to plants not involved in any R&D

ドキュメント内 Juita Binti Mohamad (ページ 190-200)

関連したドキュメント