Question

1. The following quote is excerpted from The Wall Street Journal, “Small Companies Slowly Build Momentum in the Job Market” [December 4, 2003; p. A1].

     After a long dry spell, hosts of small firms across the country are starting to take on workers again – a significant step in an economic recovery that hasn’t seen much job creation. The nation’s 23 million small businesses employ an estimated 57.1 million workers – more than half of all private-sector employees – and create more than half of the nonfarm private gross domestic product, according to the Small Business Administration.

     A wave of small-business hiring could help sustain consumer confidence and tide the economy over until larger companies regain the will to significantly boost payrolls—and begin restoring the 2.4 million jobs lost nationwide since the recession began in March 2001.

     Some large companies, benefiting from rising orders for consumer products and stronger business spending, have added people as well. November payroll numbers, drawn mainly from bigger companies, and due out Friday ... don’t detect the degree of activity by small businesses. Monthly employment data aren’t broken down by size; job-creation data specifically about smaller firms are available only annually. And employment data, drawn from surveys of established companies, often miss small start-ups. During the 20 months after the 1991 recession ended, the government said that the economy generated 303,000 jobs. The number was eventually bumped up to 663,000 because the initial surveys included only established companies.

The Bureau of Labor Statistics (BLS) publishes two main surveys on employment statistics. First, describe these two surveys and how they differ; explain the potential biases in the two surveys that could cause them to give divergent results on job growth during the beginning of an economic expansion. Second, the establishment survey data is used in computing labor productivity. There was a reported 9.4% year-to-year rise in output per worker per hour in the 3rd quarter 2003. How is labor productivity measured and why should you be suspicious that such a spectacular increase actually occurred? What factors could cause a rise in average labor productivity? Carefully explain.

Answer 1

Basic method for calucation of labour productivity is, you divide the total output by total number of labour hours.

As expalined in the above para the labour productivity increased 9.4% Y-to-Y due to rise in the overall productivity.

However the reliabilty of the data is a big question over here and there are various reason why this increase in productivty is supsicous. Following are the two main points that raised the doubt on accuracy of productivity data.

1. Firstly, the number of employees to be included in the calculation of labor productivity

2. The “dimensionality” of working time fund and its influence exerted on the level and dynamics of labor productivity

The first issue relates to the fact that it is customary to calculate labor productivity of only the so-called core production workers, i.e. those employed in the private sector with recording of data auxiliary production owned by big conglomorates. This calculation does not include a significant group of workers of “other productions” working mainly in unorganised small farm sector. This poses a number of methodological and practical difficulties in calculating labor productivity.

Labour input is measured by constructing indexes of hours worked. The total hours worked in any period is calculated as the product of the number of employed workers and average hours worked. An index of hours worked is preferable as a measure of labour input than simply the number of employed workers, as hours worked captures changes in overtime worked, standard weekly hours, leave taken, and changes in the proportion of part-time employees.11 Due to limitations in the hours worked data, the hours worked series is reported only in index form. One advantage of this is that over- (or under-) reporting of hours worked is less important if the data is in index form than if it is in levels form; often we only care about the changes in the series, and therefore the index form is entirely adequate. The employment estimates used to derive hours worked comprise all labour engaged in the production of goods and services: • civilian wage and salary earners; • employers; • self-employed persons; • persons working one hour or more without pay in a family business or on a farm;

The annual figures are simple arithmetic averages of observations on employment levels during the year. Total average hours worked are reported for each month, from the Labour Force Survey (LFS). However, estimates are available only for the mid-month of each quarter for individual industries and the market sector. The average hours worked series relate to a fortnight in the first half of the month. There are several problems with the series, including the following. • The design of the LFS means that the two week reference period is representative of only one week of each month. Thus, there are only 12 weeks of the year where total hours worked are actually observed. • Hours worked reported in the reference period may not be representative of the quarter due to holidays. Consider the following. In Australia, many people take leave in January (which is summer). The average hours worked recorded in February is then likely to overstate the average level of hours worked for the quarter. This is the main reason that the ABS presents the hours worked estimates in index form. • Hours worked in the reference period may reflect the changing incidence of public holidays. ``Calendar corrections" do not entirely remove this problem. Therefore, only the estimates of average hours worked in the mid-month of each quarter are used to derive the estimates of hours worked for ``all industries". • Details of the hours worked by members of the Australian defence forces are not available from the Household Labour Force Survey. The average weekly hours worked of civilian employees is used to proxy their hours worked per week.

So from the above analysis it is very clear that calculation of labour productivity is very complex and challaning task and the reliability of such data is always is question.