Relationship Between Labor Market Policies and Labor Productivity

[1]

 While it is commonly acknowledged that the United States has a significant lead among other developed countries in terms of labor productivity, or in other words, gross domestic product (GDP) per aggregate hour worked, little is known about the main drivers of this productivity gap. While Canada is considerably similar to the United States, the productivity differences between the two economies tell two different stories. Figure I, according to Data from the OECD, depicts the average productivity levels of developed countries on from 2000 to 2015 with a solid line. The United States, The United Kingdom, and Canada’s longitudinal data is also shown for the same period to illustrate the labor productivity gap between the United States and other developed countries. Labor productivity is a major component of a country’s economic performance and is a driver of national living standards. Labor productivity growth has slowed dramatically during the global financial crisis of 2008 according to OECD data, OECD countries have also seen lower productivity growth since the 2008 financial crisis on average. The weak post-crisis productivity growth performance is limiting the opportunity for long-term economic well-being. I predict that there are policies that can improve the allocation of scarce resources such as labor and capital which are crucial to maximizing productivity growth in the labor market. While a number of factors have been found to relate to the growth or slowdown of productivity in various countries, I will examine the relationship between labor market policies and labor productivity.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

 A nation’s labor market regulations and labor institutions have implications for macroeconomic outcomes, such as labor productivity. Regulations are rules enforced by the state that constrain the actions of economic agents, typically through the imposition of standards, such as, minimum wage regulations, which constrain the lowest possible wages paid for labor, or firing regulations, which constrain the ease at which economic agents can dismiss their employees. While regulation is essential to an effective working of a market economy and is a key function of the state, it does still, however, impose costs on the private sector and there is a risk of excessive regulation where the additional costs of compliance exceed additional benefits to the economy. As research by Crafts (2006) shows Endogenous growth theory predicts that total factor productivity growth will decrease if regulation stunts investment and innovation.  Data on world economic freedom index by the Fraser Institute Index takes into account many different labor market regulations in their indicators such as, hiring regulations and minimum wage, hiring and firing regulations, centralized collective bargaining, hours regulations, mandated cost of work dismissal, and conscription. I will examine the relationship between labor market regulations and labor productivity among a sample of 36 developed countries.

 I believe free labor markets will be conducive to higher levels of labor productivity in developed economies. I will determine empirically, the importance of labor market institutions on a developed country’s labor productivity levels.  Through a regression analysis I will be able to interpret labor market regulations relationship to a countries labor productivity from 1990 to 2015. I will then interpret the relationship between the differences in labor policies between the US and Canada to examine the determinants of the US-Canada labor productivity gap, whereas there is abundant literature on the European-American labor productivity gap but not as much on the difference present in North America. While there is existing literature regarding productivity performance, I will examine the linkage between a nation’s labor regulations with their labor productivity. Reforms that promote growth through reducing skill mismatch are essential to enabling innovative firms to attracting talent and the capital they need to grow. Inefficient labor market regulations and policies allow old unproductive firms to absorb valuable resources and talent from the economy that could be distributed or allocated in more productive ways. I predict that more lenient labor market regulations, or less employment protection legislation should result in higher productivity for developed economies.

II. Literature Review

It is clear that there is abundant research on the determinants of productivity growth among various countries. However, the role of labor market institutions and regulations in driving or stunting productivity performance has yielded conflicting results. Whereas the current literature is fragmented, with different determinants of productivity growth, different populations studied, and ambiguous results. This paper further analyzes the relationship of labor market regulations and labor productivity growth among OECD countries to deliver more concrete evidence of labor market regulation’s impact on labor productivity. Through disentangling different types of labor market regulations effects on labor productivity levels, I will contribute to existing research on productivity determinants and labor market regulations.

 One key difference between other studies and this study is the measure of productivity. I intend to analyze productivity with the labor productivity measure, which represents the volume of output produced in terms of GDP per unit of labor input in terms of aggregate hours worked. This ratio is largely impacted by the presence of capital inputs, tangible and intangible, and technical efficiency.[2] Whereas total factor productivity (TFP) measures both the labor and capital input that contributes to efficiency, labor productivity just measures output per labor input. There is empirical evidence that information and communication technology investment is a main driver of TFP growth (Cardona et all, 2013), and it is widely attributed to the gap of productivity between the U.S. and other developed countries. However what is interesting is that there is strong evidence from OECD countries that high product market regulations in Europe contributed to weak total factor productivity growth in the 1990s (Scarpetta & Nicoletti, 2003).  These studies all research different determinants of total factor productivity, which measures capital input as well as labor input in productivity. I will study labor productivity instead of total factor productivity, to separate labor input’s impact on productivity from physical capital inputs productivity impact.

Some literature regarding productivity performance only focuses on the performance of one country, rather than examining productivity growth differentials between countries. Studies on public regulations, such as Occupational Safety and Health Administration (OSHA) and the Environmental Protection Agency (EPA) in the United States (Gray, 1987; Haveman & Christainsen, 1981) found that regulations contributed to sluggish productivity growth in the manufacturing sector and in the labor productivity growth in the 1970s in the United States. Another study by Carl Magnus Bjuggren, in 2015, found that more flexible employment protection legislation in the Swedish labor market resulted in higher labor productivity. One study found that minimum wage increases may be related to increased productivity of individual workers, with evidence from one large United States retailer (Coviello et al., 2019). Whereas these studies only consider a select country, this paper provides a more concise evidence of labor market regulations impact on labor productivity growth through looking at all developed OECD countries with their various labor market institutions and productivity performance. 

 When studying papers regarding labor productivity growth and labor market regulations the evidence has been contradictory. According to one study, labor market regulations measures through employment protection has a net zero effect on productivity growth (Autor et all, 2007). Another paper found that OECD countries with relatively regulated labor markets, such as Nordic European countries, experience higher labor productivity growth (Storm & Naastepad, 2009). Research by Acemoglu and Shimer in 2000 suggests that unemployment insurance has a positive relationship with productivity through allocating workers to more productive activities and incentivizing the creation of these positions by firms. In 1999, Nickell and Lanyard’s research showed stricter employment protection in OECD labor markets had no correspondence to negative productivity growth. Most recently, there has been empirical evidence that shows that mandatory dismissal regulations have a negative impact on productivity growth, according to data from 1982 to 2003 (Bassinini et al, 2014).

The contradictory evidence proves there is further analysis needed in the relationship between labor market regulations and productivity. Whereas many industrial countries have well developed legal frameworks, high levels of human capital, and are open to trade; it is clear that the role of labor market regulations play a role in explaining the divergence of labor productivity in the economies of OECD countries.

Research by Andrew Sharpe grapples with the determinants of the US-Canada labor productivity gap, illustrated in figure I. One paper by Sharpe examines sectoral contributions to the US-Canada productivity gap, and argues that the slowdown in manufacturing sector growth alone accounts for the weaker productivity performance of Canada relative to the States. (Sharpe, 2010a). However in a 2004 study, it was found that a decline in both the manufacturing and service sectors of the economy explain the poor productivity performance of Canada relative to the United States. (Tang & Wang, 2004). In another study in 2003, Sharpe argues that productivity growth is an essential component in the growth of living standards and GDP per capita. He provides explanations for the productivity gap between Canada and the United States, stating that lower capital intensity, research & development expenditures, and human capital are the main drivers of productivity divergence between the two nations. While human capital is similar between both countries, the United States has the advantage of the high quality of their research universities, proportionally the United States has more world class university researchers than Canada (Sharpe, 2003). In another study involving Sharpe, states that the role of information and communications technology (ICT) investment was a key driver of the US-Canadian manufacturing productivity gap in the second half of the 1990s. (Bernstein et al., 2002). While research by Andrew Sharpe and others has contributed great evidence about the determinants of the productivity gap between Canada and the United States, they do not account for differences in regulatory environments in the two countries, as it is believed that labor and product market regulation is more stringent in Canada, and environmental regulation is considered by many to be more stringent in the United States, Sharpe states that it is unlikely that regulatory differences accounts for significant differences in the productivity gap. (Sharpe, 2003). Another study in 2010 from Sharpe analyzes the market-oriented policy of Canada and its potential to spur productivity growth in the Canadian economy. He argues that the overall magnitude effect of liberalization of markets in countries with existing high levels of economic liberalization, such as Canada, is not particularly large, however it is non-negligible. (Sharpe,2010b). Therefore the small effects of higher economic liberalization may pose diminishing productivity returns to market liberalization.

Other research states that the dominant driver of positive multifactor productivity growth in Canada is investment in technology, whereas the dominant driver of productivity growth in the United States is actually technological innovation and “organizational restructuring”. (Baldwin & Wulong, 2007). Daniel Trefler found empirically that there were productivity gains for both the United States and Canada as result of the North American Free Trade Agreement (Trefler, 2004). Another study by Sharpe found that relatively more competitive or less regulated markets drive innovation and productivity growth as well as positively effecting income distribution, and poverty. (Sharpe & Currie, 2008). This study promotes policies focused on increasing competition in markets as a way of increasing long term productivity for a short-term reallocation cost, and a possible way to reduce absolute poverty through distribution of income, seen in the form of lower prices from increased competition.

I will contribute to existing literature regarding the Canadian-American labor productivity gap and its determinants by examining labor market regulation data from developed countries including the U.S. and Canada.  I intend to contribute more research empirically to understand the determinants of labor productivity by investigating the relationship between the labor market of developed countries and their productivity levels.

Data

My research utilizes panel data for thirty-six OECD member countries from 1990 to 2016 from three data sources, the OECD, The Fraser Institute, and the World Bank. The OECD member countries included in my study are Australia, Austria, Belgium, Canada, Chile, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, South Korea, Latvia, Lithuania, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States.

I use longitudinal OECD data for labor productivity, investment, Employment Protection Legislation (EPL), expenditures on labor market programs (LMP), and net openness to trade for all included countries. Labor productivity is measured as the annual gross domestic product (GDP) per aggregate hours worked in a domestic economy. Investment is measured as the total percentage of GDP attributed to total investment in the economy by the OECD. The level of employment protection legislation is measured by the OECD for individual regular non-temporary contracts incorporating 8 data items measuring regulations for individual dismissals. EPL is measured from zero which represents least restrictions regarding individual dismissals, to six which represents most restrictions regarding dismissals of individuals. LMP is the measure of total expenditure as a percentage of GDP on labor market programs in a given country, provided by the OECD. In addition, I use OECD data to measure the net openness to trade of a country, which the OECD defines as the net exports of goods and services which is a change in ownership of resources and services between one economy and another, measured in million USD and percentage of GDP for net trade. This measure also includes barter transactions or goods exchanged as parts of gifts or grants between residents and non-residents.

World Bank Data is also included to compile control variables for human capital, which I measure through enrollment rate in tertiary and secondary education as a percentage of school-age population. I also measure innovation with data on gross domestic expenditures in research and development for each given country from the World Bank. This measure includes both capital and current expenditures in business enterprise, government, higher education, and the private non-profit sector. In addition I use data regarding the Agricultural, Services, and Industry sectors of the economy provided by the World bank, which are measured as percentage of total GDP attributed to that sectors economic activities. The agricultural sector corresponds to ISIC divisions 1-5 and includes forestry, hunting, and fishing as well as the cultivation of crops and livestock production. The services sector corresponds to ISIC divisions 50-99 and includes value added in services, such as education, health care, and financial services.. The value-added in the service sector is the net output of the services sector after subtracting the value of intermediate inputs. The Industry sector corresponds to ISIC divisions 10-45 and includes the manufacturing sector. This includes mining, manufacturing, construction, electricity, water, and gas industries. The value added in this sector is also the net output of a sector after adding up all outputs and subtracting intermediate inputs. For my study of the three primary sectors, the value added in each sector and the division of sectors is determined by the International Standard Industrial Classification (ISIC).

My third source of data is the Fraser Institute, which measures the economic freedom index. The economic freedom of the world index by the Fraser institute is measured covering five main areas; Size of Government, Legal System and Security of Property Rights, Sound Money, Freedom to Trade Internationally and Regulation. One sub-indicator of the regulation component of the Economic Freedom Index is Labor Market Regulations. This indicator is composed of regulations regarding minimum wages, dismissal, centralized wage setting, collective bargaining coverage, and conscription. It measures the extent that these constraints upon economic freedom are present, therefore in order to earn higher ratings, a country must allow free markets to determine wages and establish hiring and firing conditions. Each component and sub-component are measured on a scale from 0 to 10, whereas a score of 10 would represent an unregulated or constrained labor market.

In addition, I disentangle the individual effects of the different Labor Market Regulations captured in this component by breaking down the LMR indicator into its sub-indicators of hiring regulations and minimum wage; centralized collective bargaining; hours regulations; conscription; Mandated cost of worker dismissal; and hiring and firing regulations.

Hiring regulations and minimum wage measures whether fixed term contracts are prohibited for permanent tasks; the maximum cumulative duration of fixed-term contracts; and the ratio of minimum wage for a first time employee to the average value added per worker. An economy is assigned a score of 1 if fixed-term contracts are prohibited for permanent tasks and a score of 0 if they can be used for any task.[3]

Centralized Collective Bargaining is based on the Global Competitiveness Report Question which asks if wages are set by a centralized collective bargaining process or up to each individual company. A score of 7 is given to countries who respond that wage-setting is flexible and to the discretion of firms whereas a score of one is assigned to countries who wages are set by collective bargaining. [4]

Hours regulation is measured using five components: whether there are restrictions on night work; whether there are restrictions on holiday work; whether the length of the work week can be 5.5 days or longer; whether there are restrictions regarding overtime work; and whether the average annual paid leave is 21 working days or more. The zero-to-ten rating is based on how many of these regulations are in place, whereas no regulations in place would result in a score of ten and each regulation in place would result in a score deduction of 2.

Conscription is the data regarding the use and duration of military conscription. Countries without military conscription receive a score of ten, while countries with longer conscription periods receive lower ratings. [5]

The mandated cost of worker dismissal sub-indicator is based on the cost of advanced notice requirements, severance payments, and penalties due when dismissing a redundant worker with ten year tenure. The formula used to calculate ratings for mandated cost of worker dismissal is (Vmax-Vi)/(Vmax-Vmin) multiplied by ten, where Vi represents the dismissal cost measured in weeks of wages. Vmax is set at 58 weeks and Vmin is set at zero weeks, respectively. Countries outside the 58 max receive ratings of ten and those that are equal to the zero week minimum receive a score of zero.

The hiring and firing regulations sub-component is based on the Global Competitiveness Report which questions whether or not hiring and firing is impeded by regulations(which results in a rating of 1) or flexibly determined by employers, which results in a rating of 7. [6]

Table I shows the summary statistics for my data.

Table I. Descriptive Statistics

N

Minimum

Maximum

Mean

Std. Deviation

Labor Productivity

929

10.20

84.00

39.9658

15.86991

EPL-OECD

704

0.26

4.83

2.1464

0.79332

Labor Market Regulations – EFI

673

2.83

9.28

6.2262

1.44151

Labor Market Program Expenditure

736

0.01

7.19

1.7244

1.26993

Enrollment Rate, Secondary Education

830

4.79

121.86

54.3078

21.70354

Enrollment Rate, Tertiary Education

834

4.79

121.86

54.3346

21.55291

R&D Spending % of GDP

531

0.31

4.43

1.8471

0.98574

Trade Openness

958

8.97

224.84

43.2293

27.77827

Investment % of GDP

972

9.82

41.54

23.4020

4.09299

Agriculture % of GDP

868

0.23

17.48

2.9210

2.37385

service sector % of GDP

860

39.00

78.98

61.1590

6.60123

Industry Sector % of GDP

860

10.67

41.11

26.0217

5.27470

Leniency of Hiring regulations and minimum wage

638

1.83

10.00

6.5570

2.36682

Leniency of Hiring and firing regulations

676

1.33

8.51

4.4360

1.51704

Freedom to set wages without Centralized collective bargaining

673

1.83

8.85

6.0807

1.72056

Leniency of Hours Regulations

658

1.90

10.00

6.7736

2.15671

Leniency of Mandated cost of worker dismissal

605

1.01

10.00

7.4599

2.52293

Leniency of Conscription

682

0.00

10.00

6.3123

3.94538

 

Results

My regression model captures the effects of labor market regulations as a determinant of labor productivity in developed countries. The following model is estimated using OLS
LProdit=∝0+β1LMRit+β2LMPit+β3Invit+β4Industryit+β5Serviceit+β6Agrit+β7Tradeit+β8RDit+∈it

The equation shows the relationship between my dependent variable labor productivity, with my independent variable LMR, as well as my other control variables, labor market program expenditure, investment as a percentage of GDP, Sectoral composition (defined as industry, agricultural and service sectors share of GDP), net trade, and research and development expenditure as a percentage of GDP. The coefficients in my regression analysis, shown as Beta in the model, can be interpreted as the change in the labor productivity for a one unit change in the independent variable. Figure II is a linear model which depicts to the sixth regression or column in table II.

Table II

Dependent Variable: Labor Productivity

(I)

(II)

(III)

(IV)

(V)

(VI)

(VII)

(VIII)

(IX)

LMR-EFI

2.274***

2.568***

2.522***

1.798***

1.336***

1.823***

1.918***

2.046***

2.054***

 

(0.418)

(0.465)

(0.465)

(0.480)

(0.483)

(0.516)

(0.670)

(0.694)

(0.535)

LMP

2.268***

2.188***

1.448**

1.354**

0.375

0.151

0.195

0.347

(0.634)

(0.634)

(0.599)

(0.588)

(0.664)

(0.815)

(0.831)

(0.663)

Investment

 

-0.320*

0.448***

0.327*

0.585***

0.758***

0.769***

0.618***

(0.174)

(0.171)

(0.170)

(0.200)

(0.237)

(0.242)

(0.201)

Industry Sector % of GDP

-0.684***

-0.381*

-1.022***

-1.337***

-1.346***

-1.057***

(0.218)

(0.224)

(0.262)

(0.315)

(0.322)

(0.262)

Service Sector % of GDP

0.078

0.351*

-0.252

-0.581*

-0.616*

-0.274

(0.199)

(0.205)

(0.235)

(0.310)

(0.320)

(0.235)

Agriculture Sector % of GDP

 

-3.770***

-3.188***

-4.202***

-4.549***

-4.553***

-4.739***

(0.360)

(0.376)

(0.527)

(0.675)

(0.694)

(0.537)

Trade Openness

0.104***

0.092***

0.088***

0.073**

0.100**

(0.023)

(0.024)

(0.032)

(0.036)

(0.024)

R&D Spending % of GDP

1.984***

1.527*

1.495*

2.009***

(0.690)

(0.806)

(0.820)

(0.689)

Enrollment Rate, Tertiary Education

0.052

-0.045

(0.053)

(0.281)

Enrollment Rate, Secondary Education

0.103

Year (Linear Time Trend)

-0.227

(0.144)

N

670

532

532

513

513

424

354

345

424

Adjusted R Square

0.041

0.065

0.070

0.329

0.353

0.386

0.356

0.342

0.368

Table II shows linear regression results that utilize the Fraser Institute composite indicator of labor market regulations. We see a significant and positive relationship in the first column between productivity and labor market regulations. This shows a positive correlation of 2.274 between unregulated labor markets and labor productivity. This implies that each one unit increase in the labor market regulation index increases labor productivity by $2.27 USD per hour worked. On table II, the sixth column has the highest adjusted R-squared of 0.386, which signifies that this regression explains 38.6% of the variation in labor productivity for developed countries. The higher R-squared in column six, relative to the other columns, suggests that this regression model has the best fit to the data. In this regression, in the sixth column, we see that labor market regulations has a positive and significant coefficient of 2.046. This implies that for every one point increase in the LMR component rating, there would be an increase of $2.04 in GDP per aggregate labor hour worked in an economy, since LMR has higher values for less regulation. Comparing the LMR coefficient across all 8 columns, we see that labor market deregulation is consistently significant and positive in relationship with labor productivity. Labor markets relatively free from constraining labor market regulations have a significant and positive impact on productivity performance according to my data.

We control for the labor productivity determinants of human capital in the form of secondary and tertiary enrollment rate as a percentage of school age population, however we observe a decrease in the adjusted R-squared when we account for these covariates. The model in column five has the highest adjusted R-squared and therefore is the model with the best fit. I interpret my regression results from this model in column five because this model has the highest adjusted R- squared or has the best fit. The covariate expenditure on labor market programs(LMP), is insignificant in the fifth regression analysis or column of table II, however Investment as a percentage of GDP is significant and positive with the coefficient of 0.585. This indicates that a one percent increase in Investment as a percentage of GDP is corresponds to a 58 cent increase in labor productivity, or GDP per aggregate labor hour worked.

Additionally when we control for sectoral composition we find that the covariates for Industry and Agricultural sectoral activities as percentage of GDP is significantly and negatively related to labor productivity. However, the covariate for service sector activities as a percentage of GDP was positive and insignificant in the sixth regression analysis of Table II. The coefficient for Industry Sector activities is -1.022. This indicates that a one percent increase in the percent of GDP that is contributed to industrial activities would correspond to a loss of $1.02 USD in GDP per aggregate hours worked. The coefficient for the agricultural activities covariate is -4.202, which indicates that there is about $4.20 lost in GDP per aggregate hour worked in the economy, that corresponds to a one percent increase of agricultural sector activities per GDP in developed economies.

The Trade Openness or Net Trade covariate has a positive and significant coefficient of 0.092. This implies that increasing net trade (exports – imports), is significant to increasing labor productivity. The coefficient indicates that one dollar increase in net trade, corresponds to a 9 cent increase in GDP per aggregate hour worked in a developed economy. Including the covariate for trade openness does not change the interpretations of LMR, my main variable of interest. However, trade openness may exhibit reverse causality as net trade may be determined by productivity, instead of productivity being determined by net trade. Due to the possibility of reverse causality for this covariate, the interpretation of this coefficient may be biased.

Research and Development expenditure as a percentage of GDP is also positive and highly significantly related to labor productivity in developed economies. The coefficient for the R&D covariate is 1.984. This indicates that a one percent increase in R&D spending as a percent of GDP, corresponds to a $1.98 increase in labor productivity or GDP per aggregate labor hour worked in a developed economy.

This study also controls for the linear time trend of my data in the ninth column of my table. The results indicate that there is no major change to the coefficient of my main variable of interest, Labor Market Regulations. Additionally,  the coefficient for year or linear time trend is insignificant, which indicates there is not a significant relationship between labor productivity and the linear time trend of my empirical data. Therefore, I can conclude that my results are not attributed to the linear time trend, as there are no significant changes to my coefficient for my main variable of interest and the relationship of labor productivity and the linear time trend of my data is not significant.

Table III

Dependent Variable: Productivity

 

(I)

(II)

(III)

(IV)

(V)

(VI)

(VII)

Leniency of Hiring and firing regulations

-0.325

2.562***

2.598***

1.628***

2.624***

2.967***

2.576***

(0.441)

(0.432)

(0.458)

(0.443)

(0.471)

(0.483)

(0.270)

Firms Freedom to set wages (Centralized collective bargaining)

-5.260***

-5.260***

-4.150***

-4.661***

-4.739***

-1.483***

(0.403)

(0.404)

(0.401)

(0.401)

(0.398)

(0.274)

Leniency of Hours Regulations

-0.075

-0.297

-0.682**

-0.394

-0.047

(0.320)

(0.299)

(0.300)

(0.315)

(0.159)

Leniency of mandated cost of worker dismissal

1.618***

1.138***

1.209***

-0.524***

(0.199)

(0.214)

(0.214)

(0.125)

Freedom from Conscription

0.846***

0.796***

0.636***

(0.163)

(0.163)

(0.100)

Leniency of Hiring & Minimum Wage Regulations

-0.706***

 

 

(0.254)

Real Minimum Wage (In 2017 constant prices at 2017 USD PPPs)

 

3.416***

(0.190)

LMP

-0.094

-0.345

-0.358

-0.357

-0.079

-0.082

0.097

(0.671)

(0.566)

(0.571)

(0.532)

(0.519)

(0.515)

(0.263)

Investment

0.581***

0.969***

0.963***

0.778***

0.883***

0.892***

0.314***

(0.203)

(0.173)

(0.176)

(0.166)

(0.162)

(0.161)

(0.091)

Industry Sector % of GDP

-0.718***

0.381*

0.408

0.201

0.323

0.392*

0.928***

(0.251)

(0.228)

(0.256)

(0.241)

(0.235)

(0.234)

(0.134)

Service Sector % of GDP

0.180

1.244***

1.264***

0.886***

0.775***

0.859***

0.409***

(0.208)

(0.193)

(0.211)

(0.203)

(0.198)

(0.198)

(0.110)

Agriculture % of GDP

-3.588***

-1.942***

-1.920***

-1.968***

-2.094***

-2.081***

-0.578***

(0.529)

(0.463)

(0.476)

(0.447)

(0.434)

(0.431)

(0.212)

Trade Openness

0.113***

0.130***

0.130***

0.106***

0.088***

0.087***

0.118***

(0.024)

(0.020)

(0.020)

(0.019)

(0.019)

(0.019)

(0.009)

R&D Spending % of GDP

2.168***

-0.293

-0.280

0.674

0.922

1.021*

-1.786***

(0.724)

(0.638)

(0.642)

(0.611)

(0.595)

(0.591)

(0.344)

N

424

424

423

421

421

421

300

Adjusted R Square

.369

0.552

0.551

0.612

0.635

0.641

0.926

Table III shows the sub-indicators of the Labor Market Regulation component of the Economic Freedom index. Through introducing the sub-components one at a time, and holding the other control variables constant, I disentangle the effects of each individual type of labor market regulation. The sixth column shows the OLS regression with the highest adjusted R square using all six sub-components with an adjusted R squared being 0.641. When we substitute the leniency of hiring and minimum wage regulations sub-component for OECD data for Real minimum wage in 2017 U.S. Dollars Purchasing Power Parity (PPP), we get a higher adjusted R squared of 0.926.

Mandated cost of worker dismissal has a negative and significant coefficient of -0.524, indicating that an increase of one point on the ten point scale of cost of dismissal, signifies a 52 cent loss in GDP per aggregate labor hour worked in a developed countries economy. Conscription has a significant coefficient of 0.636 which indicates that for every one point increase, on the ten point scale of freedom from conscription, coincides with a roughly 64 cent increase to a developed nation’s GDP per aggregate labor hour worked in the economy. Additionally, a one point increase on the ten point scale of hiring and firing sub-component scale corresponds to a $2.58 loss to a developed economies GDP per aggregate labor hour worked in the economy, as shown by the positive and significant coefficient in the first row, last column, 2.576. It is also noted that hours regulations have an insignificant relationship with labor productivity.

Firms freedom to set wages, in absence of centralized collective bargaining, has a significant and negative relationship with labor productivity performance, reflected by the coefficient -1.483. This implies that centralized collective bargaining is conducive to increasing productivity performance, rather than decreasing it. In addition, minimum wage regulations has a significant and negative relationship with labor productivity.

The coefficient for leniency in minimum wage regulations in the sixth column is -0.706, which suggests a standard for minimum wages is also conducive for productivity performance in developed countries. However, the variables mandated cost of worker dismissal, conscription, and hiring and firing regulations have a significant and positive relationship with labor productivity, which shows they have a negative relationship with labor productivity performance. In the seventh column or regression analysis, we substitute the Fraser institute’s subcomponent indicator of hiring and minimum wage regulations with the real minimum wage data from the OECD in the units of 207 U.S. Dollars PPP. This regression is a better fit of the data when we substitute the subcomponent for real minimum wage data, reflected by the Adjusted r-squared increasing to 0.926 (from 0.641). The coefficient for real minimum wage in 2017 USD PPP is highly significant and positive at 3.416. This implies that for every one dollar increase, in 2017 US dollars PPP, in the minimum wage of a developed economy, is correlated to a $3.42 increase in labor productivity. Whereas the United States productivity level was 63.5 GDP in USD, a 1$ increase in minimum wage regulations in 2017 USD PPP, corresponds to a $3.42 increase in labor productivity or labor productivity growth of 5.4%.

Robustness Test

In order to test the robustness of my results above, I substitute the Fraser Institute’s measure of labor market regulation for the OECD’s measure of employment protection legislation (EPL) as my independent variable of interest. Table IV shows the OLS regression results analyzing the effects of employment protection legislation and other control variables with the dependent variable, labor productivity. It is clear initially that employment protection legislation has a negative correlation with labor productivity from the coefficient of -3.181. This corresponds to a GDP loss of 3 dollars and 18 cents per aggregate labor hour, from a one point increase on the six point scale measuring the strictness of employment protection legislation in the labor market. This confirms the robustness of my results in previous tables, as the results show a negative relationship between labor markets constrained with regulations when using measures from two different third party sources, the Fraser Institute and the OECD. Therefore results would be stable when substituting data from the Fraser institute for the OECD for labor market regulations.

I acknowledge that there may be reverse causality for my covariates, however I do not believe that my variable of interest LMR exhibits reverse causality as the degree of regulation in a labor market is not determined by productivity. It is safe to assume that government regulations are not usually created for sake of efficiency or productivity, but rather to enforce a degree of equity and safety in society. Therefore, the covariate interpretations should only be taken with a grain of salt, as they may exhibit reverse causality, but my interpretation of my main variable of interest for the purpose of this study (LMR) is still significant, and unlikely exhibits reverse causality.

Table IV

Dependent Variable Labor Productivity

 

(I)

(II)

(III)

(IV)

(V)

(VI)

(VII)

(VIII)

EPL-OECD

-3.181***

-3.083***

-3.470***

-6.13***

-6.079***

-5.095***

-5.162***

-5.203***

(0.679)

(0.816)

(0.806)

(0.868)

(0.848)

(1.092)

(1.444)

(1.469)

LMP

0.712

0.399

0.832*

0.874**

0.308

0.047

0.154

(0.532)

(0.527)

(0.448)

(0.438)

(0.703)

(0.915)

(0.941)

Investment (% of GDP)

-0.84***

0.125

0.026

0.622**

0.826***

0.833***

(0.184)

(0.173)

(0.170)

(0.250)

(0.299)

(0.304)

Industry Sector % of GDP

-1.394***

-1.166***

-1.617***

-1.859***

-1.875***

(0.204)

(0.205)

(0.285)

(0.334)

(0.341)

Service Sector % of GDP

-0.752***

-0.548***

-0.78***

-1.01***

-1.04***

(0.198)

(0.198)

(0.257)

(0.326)

(0.335)

Agriculture % of GDP

 

-4.063***

-3.585***

-5.195***

-5.869***

-5.909***

(0.275)

(0.286)

(0.583)

(0.724)

(0.744)

Trade Openness

0.121***

0.083***

0.063

0.061

(0.025)

(0.029)

(0.038)

(0.039)

R&D Spending % of GDP

1.199

0.729

0.81

(0.840)

(1.076)

(1.101)

Enrollment Rate, Tertiary Education

-0.004

0.446

(0.059)

(0.605)

Enrollment Rate, Secondary Education

-0.452

(0.605)

N

692

530

530

487

487

317

265

258

Adjusted R Square

0.029

0.027

0.063

0.447

0.472

0.445

0.412

0.410

In the fifth column of table IV, we see the robustness test regression result with the highest adjusted R square of 0.472, as we control for other variables. In this regression, we see that EPL has a significant coefficient of -5.203, which indicates that a one point increase in EPL would correspond to a loss of about $5.20 cents of GDP for every aggregate man hour worked in the economy. This suggests that EPL has a significant and negative relationship with labor productivity as we continue to control for more variables.

Figure II.

 I test for heteroscedasticity in the above heteroscedasticity scatterplot. As there is no evident pattern in the data plots, I can assume that there is no heteroscedasticity in my data, and therefore I can justify using homoscedastic error terms in my result.

Conclusion

My empirical evidence shows that labor markets that are relatively more free of labor market regulations and employment protection legislation in developed economies have a strong and significant relationship with positive labor productivity levels. Therefore, empirically it concludes that strict labor market regulations in the labor market are not favorable to positive labor productivity levels in developed economies.  In table II, I have shown that strict labor market regulations have a negative correlation alone with labor productivity performance of developed economies. When controlling for other relevant variables my OLS regressions show that the more lenient labor market regulations in a developed economy, the more conducive are the conditions for increasing labor productivity. Table III disentangles the separate effects of different labor market regulations which accounts for positive and negative relationships between certain regulations and productivity performance. While I found centralized collective bargaining coverage and minimum wage to have a positive relationship with labor market regulations in my data, Labor market regulations had an overall negative effect on productivity performance for developed economies in my study. This is due to the largely negative effects on productivity caused by mandated cost of worker dismissal, conscription, and hiring & firing regulations. It is therefore expected from my results that labor market regulations concerning wages of workers is conducive and correlated with increasing labor productivity levels, as seen in my regression results for the sub-component indicators of leniency of hiring and minimum wage regulations and freedom to set wages without centralized collective bargaining. However, the largely negative effects of leniency of mandated cost of worker dismissal and leniency of firing regulations offset the positive effect of wage regulations, therefore making the net effect of labor market regulations on labor productivity negative.

Labor market regulations concerning wages may be conducive to productivity as they may increase incentives and motivations for existing workers, and possibly attract a more skilled workforce through higher wages. It was found from evidence from a large United States retailer that a one dollar increase in minimum wage is correlated to a 4.5% increase in labor productivity (Coviello et al., 2019). This is similar to and supports my findings in my linear regression analysis that a one dollar increase in minimum wage corresponds to a 5.4% increase in labor productivity. According to the research from a large U.S. retailer, higher minimum wages may lessen the incentives for workers to exert more effort, but on the other hand, workers are made better off, and therefore may care more about the retention of their employment, and in consequence their job performance. (Coviello et al., 2019).

It is informative to consider the impact to which the effect labor market regulation, on labor productivity in developed countries, estimated above, can explain the Canada-US productivity gap. In 2016, Canada had a Labor Market Regulation component rating by the Fraser Institute of 8, compared to the United States’ rating of 8.83. Higher ratings reflect relatively unconstrained or unregulated free labor markets. In the same year, Canada experienced a labor productivity level of 42.84 according to OECD Data, in comparison to the labor productivity level of 53.39 in the United States. According to my significant and positive coefficient of 1.918 for Labor market regulations in the sixth regression analysis of Table II, increasing the rating of LMR in a country by one would correspond to a $1.92 increase in a nation’s productivity level. Given the difference in regulations between Canada and the US, the estimates in this study suggest that Labor market regulations can explain roughly 15.1% of the difference in labor productivity levels between the United States and Canada in 2016.

Supported by research contained in my literature review, the relationship of labor market regulations (LMR) is best interpreted as small, but nonnegligible determinant of the United States-Canada labor productivity gap. In future research, we could incorporate  measures of human resource management and geographic labor mobility as control variables. These may be determinants of productivity and could explain more of productivity differences between the United States and Canada, however for the purpose of the study I was not able to find reliable measures of these covariates. It is important to realize there are potentially diminishing productivity returns to the market liberalization of already relatively liberal, developed economies. However, in the argument of productivity-centered economic policy, it is also important to promote equitable distribution of productivity gains in order to capture public support. Increasing competition intensity through the removal of formal barriers in the form of market regulations has the potential to reduce absolute poverty through savings passed on to consumers by fostering a more competitive business environment. Productivity is a significant contributor to growth in GDP per capita, and therefore national living standards. However, labor productivity is also affected by various, small, and incremental changes, so it is crucial to not underestimate, as well as over-emphasize the effects of labor market deregulation on productivity performance of developed economies.

Appendix

Table A.I

 

 

Bibliography

“Approach – Fraser Institute.” FraserInstitute.org. 12/13/ 2017. Web. 02/22/19 <https://www.fraserinstitute.org/economic-freedom/approach?active-tab=0>.

Acemoglu, Daron, and Robert Shimer. “Productivity Gains From Unemployment Insurance.” European Economic Review 44.7 (2000): 1195-1224.

Alfaro, Laura, Sebnem Kalemli-Ozcan, and Selin Sayek. “FDI, Productivity and Financial Development.” World Economy.

Andrea Bassanini, Luca Nunziata, and Danielle Venn. “Job Protection Legislation and Productivity Growth in OECD Countries.” Economic Policy 24.58 (2009): 349-402. 

Auer, Peter, Janine Berg, and Ibrahim Coulibaly. “Is a Stable Workforce Good for Productivity?” International Labour Review 144.3 (2005): 319-43. 

Autor, David, William Kerr, and Adriana Kugler. “Does Employment Protection Reduce Productivity? Evidence from the US States.” The Economic Journal (2007).

Belot, Michèle, Jan Boone, and Jan Van Ours. “Welfare‐improving employment protection.” Economica 74.295 (2007): 381-396.

Bernstein, Jeffrey & Harris, Richard & Sharpe, Andrew. “The Widening Canada-US Productivity Gap in Manufaturing,” International Productivity Monitor, Centre for the Study of Living Standards. 2002. vol. 5. pages 3-22. Fall. https://ideas.repec.org/a/sls/ipmsls/v5y20021.html.

Bjuggren, Carl Magnus. “Employment Protection and Labor Productivity.” Journal of Public Economics 157 (2018): 138-57. 

Buchele, Robert, and Jens Christiansen. “Employment and Productivity Growth in Europe and North America: The Impact of Labor Market Institutions.” International Review of Applied Economics 13.3 (1999): 313-32. 

Cardona, M; Kretschmer, T; Strobel, T. “ICT and Productivity: Conclusions from the Empirical Literature.” Information Economics and Policy 25.3 (2013b): 109-25. 

Cardona, M., T. Kretschmer, and T. Strobel. “ICT and Productivity: Conclusions from the Empirical Literature.” Information Economics and Policy 25.3 (2013a): 109-25. .

Cette, Gilbert, Jimmy Lopez, and Jacques Mairesse. Product and Labor Market Regulations, Production Prices, Wages and Productivity. 2014. 

Crafts, Nicholas. “Regulation and Productivity Performance.” Oxford review of economic policy 22.2 (2006): 186-202.

Francisco Alcalá, and Antonio Ciccone. “Trade and Productivity.” The Quarterly Journal of Economics 119.2 (2004): 613-46. 

Freeman, Richard. Labor Regulations, Unions, and Social Protection in Developing Countries Market Distortions Or Efficient Institutions?.

Gray, Wayne Burger. “The Cost of Regulation.” The American economic review 77.5 (1987): 998-1006.

Gwartney, James. Economic Freedom of the World. Fraser Institute, 2018. Web.

Haveman, Robert H., and Gregory B. Christainsen. “Public Regulations and the Slowdown in Productivity Growth.” American Economic Review 71.2 (1981): 320-25.

Javorcik, Beata Smarzynska. “Does Foreign Direct Investment Increase the Productivity of Domestic Firms?” The American economic review 94.3 (2004): 605-27.

Mary, O’Mahony, and Nicholas Crafts. “A Perspective on UK Productivity Performance.” Fiscal Studies 22.3 (2001): 271-306.

Nickell, Stephen, and Richard Layard. “Labor market institutions and economic performance.” Handbook of labor economics 3 (1999): 3029-3084.

OECD (2018), GDP per hour worked (indicator). doi: 10.1787/1439e590-en (Accessed on 07 November 2018)

OECD. OECD Compendium of Productivity Indicators 2012. 1. Aufl. ed. FR: OECD Paris, 2013.

Pagés, Carmen. “A Cost-Benefit Approach To Labor Market Reform.” Economic Review 89.2 (2004): 67-85.

Robinson, James Alan, Daron Acemoglu, and Simon Johnson. Institutions as the Fundamental Cause of Long-Run Growth. 2004.

Scarpetta, Stefano, and Giuseppe Nicoletti. “Regulation, Productivity and Growth: OECD Evidence.” Economic Policy18.36 (2003): 9-72.

Sharpe, Andrew. “Why are Americans More Productive than Canadians?” Centre for the Study of Living Standards. 6. (2003): 3-14.

Stansel, Dean. Economic Freedom of North America 2017. Fraser Institute, 2017. Web.

Storm, Servaas, and C. W. M. Naastepad. “Labor Market Regulation and Productivity Growth: Evidence for Twenty OECD Countries (1984-2004).” Industrial Relations 48.4 (2009): 629-54. 

Sharpe, Andrew. “Ten Productivity Puzzles Facing Researchers,” International Productivity Monitor. Centre for the Study of Living Standards.(2004). 9. 15-24. https://ideas.repec.org/a/sls/ipmsls/v9y20042.html.

Sharpe, Andrew. “Can Sectoral Reallocations of Labour Explain Canada’s Absymal Productivity Performance?.” International Productivity Monitor 19 (2010a): 40.

Trefler, Daniel. “The long and short of the Canada-US free trade agreement.” American Economic Review 94.4 (2004): 870-895.

Sharpe, Andrew, and Ian Currie. Competitive intensity as driver of innovation and productivity growth: A synthesis of the literature. No. 2008-03. Centre for the Study of Living Standards, 2008.

Baldwin, John R., and Wulong, Gu. “Long-term productivity growth in Canada and the United States.” Canadian Productivity Review Research Paper 13 (2007).

Rao, Someshwar, Andrew Sharpe, and Jeremy Smith. “An analysis of the labour productivity growth slowdown in Canada since 2000.” International Productivity Monitor 10 (2005): 3-23.

Tang, Jianmin, and Weimin Wang. “Sources of aggregate labour productivity growth in Canada and the United States.” Canadian Journal of Economics/Revue canadienne d’économique 37.2 (2004): 421-444.

Sharpe, Andrew. The paradox of market-oriented public policy and poor productivity growth in Canada. No. 2010-01. Centre for the Study of Living Standards, 2010b.

Coviello, Decio, Erika Deserranno, and Nicola Persico. “Minimum Wage and Individual Worker Productivity: Evidence from a Large US Retailer.” (2019).

[1] OECD (2018), GDP per hour worked (indicator). doi: 10.1787/1439e590-en (Accessed on 07 November 2018)

[2] OECD. OECD Compendium of Productivity Indicators 2012. 1. Aufl. ed. FR: OECD Paris, 2013. Web.

[3] A score of 1 is assigned if the maximum cumulative duration of fixed-term contracts is less than 3 years; 0.5 if it is 3 years or more but less than 5 years; and 0 if fixed-term contracts can last 5 years or more. Finally, a score of 1 is assigned if the ratio of the minimum wage to the average value added per worker is 0.75 or more; 0.67 for a ratio of 0.50 or more but less than 0.75; 0.33 for a ratio of 0.25 or more but less than 0.50; and 0 for a ratio of less than 0.25.

[4] World Economic Forum, Global Competitiveness Report. 

[5] When length of conscription is six months or less, the country earns a rating of five for conscription. If conscription has a duration of 6 to 12 months they earn a rating of 3, and if it is between 12 and 18 months they earn a rating of 1 in conscription. Any conscription greater than 18 months results in a rating of zero. Special cases include when mandated national military service includes non-military options they receive a rating of five, and when it is clear conscription is never used, but still remains possible countries still receive a score of zero.

[6] World Economic Forum, Global Competitiveness Report.