92 years of economic insights for Indiana

The IBR is a publication of the Indiana Business Research Center at IU's Kelley School of Business

Lean manufacturing: The production employment and wages connection

Assistant Professor, University of Indianapolis, School of Business

Associate Professor, University of Indianapolis

The manufacturing sector of the economy is increasingly characterized by complexity and rapid change. The inherent volatility of manufacturing is often the source of workforce apprehension, which produces a reluctance to adapt to the changing environment.

To make (or manufacture) things, it costs money—facilities, materials and labor are just three of the many “inputs” necessary—and it makes sense to add as much value to a product as possible.  Lean manufacturing has long been one of the ways to add value by improving the processes involved in the making of the product and eliminate waste.

Workplace efficiency initiatives can sometimes be misinterpreted as attempts to disrupt the workplace or undermine the social contract between the employer and employees. Within this context, the workforce can resist the adoption of lean manufacturing or other manufacturing efficiency programs. The lack of direct research to counteract this viewpoint only re-enforces its legitimacy. This article is a step toward understanding the relationship between value-adding (or lean) manufacturing and employment and wages.

This analysis is in the same line as well-established research in the area of efficiency wage hypothesis. Efficiency wage hypothesis (EWH) research examines the potential for wage premiums over market-clearing wages to attract better talent. The goal is to acquire better talent that produces productivity gains and reductions in turnover costs in excess of the increased wages.1 This analysis differs in that it does not assess the performance impacts of wage premiums of individual companies, but rather uses statewide wage averages. Using statewide averages, we examine the directional link between productivity and market-clearing wages, rather than the premium wages of individual actors within industry.

Figure 1: Comparison of value added per hour

graph

Source: U.S. Census Bureau, Annual Survey of Manufactures

Table 1: Production by the numbers

In order by highest value
added per hour
Value added
per hour of work
Average wage
per hour
Indiana
production workers
Percent
of U.S.
Indiana U.S Indiana U.S
 Manufacturing $149.99 $151.50 $22.28 $22.15 349,425 4.5%
Top 5 325 Chemicals $796.16 $432.60 $27.49 $29.08 11,826 2.7%
324 Petroleum and coal products $732.75 $646.41 $41.53 $38.32 2,685 4.0%
312 Beverage and tobacco products $375.36 $534.62 $25.11 $26.91 1,482 1.8%
334 Computers/electronic products $211.17 $240.47 $22.22 $26.29 8,449 2.3%
331 Primary metals $168.37 $138.32 $31.00 $25.99 34,935 11.6%
311 Food $162.91 $123.10 $19.31 $17.65 24,357 2.2%
339 Miscellaneous $162.03 $158.97 $20.02 $19.52 17,347 5.4%
336 Transportation equipment $126.99 $157.19 $23.23 $26.82 90,981 9.0%
333 Machinery $126.73 $146.77 $21.83 $23.39 24,137 3.7%
322 Paper $107.39 $155.49 $21.71 $25.44 6,989 2.7%
327 Nonmetallic minerals $98.35 $113.58 $22.22 $21.53 10,592 3.9%
332 Fabricated metals $96.74 $91.41 $20.29 $21.29 42,549 4.2%
335 Electrical equipment/components $92.69 $135.27 $20.23 $21.39 4,971 2.2%
Bottom 5 326 Plastics and rubber $77.12 $94.51 $17.55 $18.75 31,166 5.6%
337 Furniture $75.61 $73.97 $17.15 $17.05 14,011 5.6%
323 Printing  $73.02 $82.03 $19.42 $20.29 10,071 3.4%
321 Wood products $56.64 $69.25 $17.23 $17.25 10,433 3.7%
314 Textile product mills $43.07 $65.48 $13.06 $15.15 1,504 1.8%

Source: U.S. Census Bureau, Annual Survey of Manufactures

Our sources include data from the U.S. Census Bureau’s Annual Survey of Manufactures (ASM). It includes statewide, industry-level data for the 50 states and the District of Columbia beginning in 1998 through 2013.2 These data include total number of employees, total number of production employees, total wages for production employees, total production hours for production employees, value added by the company, and capital expenditures by the company. Calculated variables were also added. Summary statistics are provided in Table 2. (Editor's note: Table 2 is for readers who want to see the statistical results of the number crunching. Others may want to skip this part).

Table 2: Summary statistics

Variable Number of
Observations
Mean Standard deviation Minimum Maximum
Total employees 13,767 1,551.5 23,126.7 0 395,299
Production employees 13,728 10,957.0 15,907.4 0 220,427
Total wages for production employees 13,698 365,000,000 620,000,000 0 12,000,000,000
Total hours for production employees 13,748 20,900,000 32,000,000 0 482,000,000
Production employee wages per hour (1) 13,587 17.27 5.3 0 50.35
Wage per production employee (2) 13,569 31,993 13,825.8 0 114,817
Value added 13,428 2,320,000,000 4,370,000,000 0 80,600,000,000
Value added per production hour (3) 13,310 118.0 141.2 0 2,679.9
Capital expenditures 12,313 154,000,000 335,000,000 0 6,830,000,000
Capital expenditures per production employees (4) 12,168 14,812.0 31,205.5 0 1,279,897.0
Ratio of total employees to production employees (5) 13,620 1.4 0.3 0 8.0

(1)   Calculated as: Total Wages for Production Employees / Total Hours for Production Employees
(2)   Calculated as: Total Wages for Production Employees / Production Employees
(3)   Calculated as: Value Added / Total Hours for Production Employees
(4)   Calculated as: Capital Expenditures / Production Employees
(5)   Calculated as: Total Employees / Production Employees
Source: Authors’ calculations, using U.S. Census Bureau Annual Survey of Manufactures data

The analysis used an ordinary least squares (OLS) model. Yearly binaries were created to account for yearly influences, such as the general economic climate and inflation. State binaries were created to account for state-specific influences and cost of living differences. Industry binaries were created to designate the type of manufacturing and account for the differing wage scales that exist in manufacturing due to job variation. The manufacturing designations are based on three-digit North American Industry Classification System (NAICS) codes. This analysis assumes that different manufacturing jobs require differing levels of employee skill. The more skills required for specific jobs or sectors within manufacturing, the higher the wage.3

This article attempts to determine the impact of lean manufacturing on employment and wages. Lean refers to “manufacturing that focuses on reducing or eliminating waste in all facets of the system.”4 The study uses value added per production employee hour as a proxy for lean.

As a manufacturer becomes lean, it should increase the value added it provides per production employee per hour.

Gross value added per employee has been used in prior studies.5 The variable, value added per production employee, is used as explanatory in this analysis.

Investment was a possible explanatory variable since capital investment can influence the level of value added in the manufacturing process. The ratio of total employees to production employees was also added as an explanatory variable. This variable ascertains the influence of differing employee structures.

The non-binary data was transformed to natural log form to help account for exponential effects and to aid in the interpretation of the data. As both the independent and explanatory variables of note are in natural log form, the magnitudes of the result coefficients will be elasticities. Increases in value added per employee hour (or becoming lean) have a lingering effect on wages (existence of autocorrelation). A lagged term was added to the model in order to accommodate this influence. The lagged variable is the value added per production employee term.

The results establish a strong correlation between value added and wages. Higher levels of worker productivity coincide with higher worker wages. The results of the non-binary variables are provided in Table 3. (Download the appendix for the full results.) The value added and its lag are both highly significant and positive in all four models.

Table 3: Regression results (non-binary variables)

  1 2 3 4
  Production employees Total hours for production employees Production employee wages per hour Wage per production employee
Adjusted R-square 0.747 0.919 0.850 0.996
  Coefficient Standard error Coefficient Standard error Coefficient Standard error Coefficient Standard error
Value added per production employee 0.161 0.031* 0.131 0.031* 0.139  0.006* 0.109 0.006*
Value added per production employee lagged (one year) 0.320 0.031* 0.334 0.031* 0.051  0.006* 0.064 0.006*
Capital expenditures per production employees 0.250 0.012* 0.270 0.012* 0.055  0.002* 0.075 0.002*
Ratio of total employees to production employees -0.641 0.059* -0.590 0.059* -0.080  0.011* -0.029 0.011*

* Statistically significant at the 1 percent level.
Source: Authors’ calculations, using U.S. Census Bureau Annual Survey of Manufactures data

Higher levels of value-added production are positively correlated with more production employees, more total hours for production employees, higher production employee wages per hour and higher total wages per production employee. Increasing the value added per employee is correlated with increasing levels of production employment and higher production wages. However, this study lacks the explanatory variables necessary to claim a causal link, so the focus is correlation.

The capital expenditure per production employee variable is highly significant and positive in all four models. Capital expenditures are positively correlated with more production employees, more total hours for production employees, higher production employee wages per hours and higher total wages per production employee.

Efforts in lean manufacturing are often coupled with capital expenditures. These capital expenditures, rather than diminishing the need for workers, are correlated with increasing levels of production employment and higher production wages.

The ratio of total employees to production employees is highly significant and negative in all four models. Manufacturing companies that are increasingly top heavy are correlated with fewer production employees, fewer total hours for production employees, lower production employee wages per hour and lower total wages per production employee.

The results generally suggest a positive relationship between lean manufacturing and its impact on worker welfare. Value-added enhancement is associated with higher levels of production employment and wages.

However, the advances in employment and wage growth are less than proportional to the increase in value added. A 1 percent increase in value added only results in a 0.16 percent increase in production employment, a 0.13 percent increase in total hours for production employees, a 0.14 percent increase in production employee wages per hour and a 0.11 percent increase in total production employee wages. This indicates that a smaller portion of the benefits of value-added enhancement activities are allocated to production employees.

This information is important for policy planning in the areas of business development and education. Economic developers should focus on bringing high value-added manufacturing to the state if they want long-term increases in wage and employment growth. A state with low-skill manufacturing will likely languish with slow wage growth and lower rates of employment growth. If a state wishes to bring in higher value-added manufacturing, they will also need a workforce with the required skills, so having adequate educational services to provide for this demand is important.

Notes

  1. J.E. Stiglitz, “The Efficiency Wage Hypothesis, Surplus Labour, and the Distribution of Income in L.D.C.s,” Oxford Economic Papers 28, no. 2 (1976): 185–207.
  2. U.S. Census Bureau Annual Survey of Manufactures data obtained from STATS Indiana (www.stats.indiana.edu/asm/) on November 12, 2015.
  3. S. Snell and J. Dean, “Integrated Manufacturing and Human Resource Management: A Human Capital Perspective,” Academy of Management Journal 35, no. 3 (1992): 467-504.
  4. “APICS Operations Management Body of Knowledge Framework, Third Edition.” APICS Magazine, 2015, www.apics.org/industry-content-research/publications/ombok/apics-ombok-framework-table-of-contents/apics-ombok-framework-3.11.
  5. T. Kochan, R. Lansbury, and J. P. MacDuffie, After Lean Production: Evolving Employment Practices in the World Auto Industry (Ithaca: Cornell University Press, 1997).