OLMIS

by Phoebe Colman
Published Dec-20-2011

Getting to Know the Data Oregon Size of Firm Data for 2010 Shares Remain Steady From Year to Year Definitions Count: Oregon Compared With the Nation Distributions Vary by Industry What We Still Don't Know

What does it mean to talk about business size? In public policy discussions it often means analyzing relationships between the size of a business and its impact on the economy. In the United States, a common point of debate is whether small businesses are the drivers of private-sector job creation. Underlying this debate are numerous statistical and measurement issues that make definitive answers difficult to come by.

The purpose of this article is to provide a portrait of Oregon private-sector businesses by size of firm, with careful attention to how such firms are classified and counted. The figures presented here offer details about Oregon's economy at a specific point in time, but they do not describe dynamic processes such as economic growth or job creation. Due to the nature of the data, this analysis makes no claims for or against the job-creating power of small businesses.

Size of firm data come from the Quarterly Census of Employment and Wages (QCEW) program. This program collects information from quarterly tax reports filed by employers under Oregon unemployment insurance (UI) laws. Employers subject to UI reporting requirements account for approximately 95 percent of all employment in Oregon.

Defining what counts as a "firm" is a basic methodological problem and can greatly impact analysis and conclusions. A firm is defined as a single business, either corporate or otherwise, that may consist of one or more operating establishments in Oregon. This definition is useful when discussing the impact a firm has on the statewide economy. It does have some limitations when comparing Oregon size of firm data with similar data for the nation as a whole.

This article uses the same size classes as those found in most Bureau of Labor Statistics (BLS) publications, and follows BLS methods by using March employment data from private-sector employers to classify individual firms. For example, a firm reporting seven employees in March 2010 is placed in the five to nine size class for 2010. Data on number of firms and wages at those firms are based on first quarter reports for the specified year. Although seasonal variations exist in quarterly wages, numbers of employees, and numbers of firms in each size class, their overall distribution does not vary significantly from quarter to quarter.

Tables and graphs in this article will omit size class zero. While firms reporting no employment in March represent approximately 15 percent of the total number of firms in the QCEW database, the total wages for these firms account for less than 1 percent of total wages in Oregon.

More than half of the private-sector firms in Oregon had fewer than five employees in March 2010. Firms with fewer than 20 employees accounted for 89 percent of all Oregon firms. The number of firms in each size class decreases steadily as the size class gets larger (Table 1).

Although smaller firms are more numerous, they generally employ a smaller share of the total number of workers than larger firms. For example, the 89 percent of Oregon firms with fewer than 20 employees accounted for about 27 percent of private-sector employment in 2010. By comparison, firms with 250 employees or more represented less than 1 percent of the total number of firms but employed about 35 percent of Oregon workers.

The distribution of wages paid in Oregon follows a pattern similar to that of workers. The share of total wages for a given size class is generally within about 2 percentage points of the share of total employment for that size class. Among firms with fewer than 250 employees, the share of total wages falls below the share of total employment for each size class. The share of total wages is slightly higher than the share of total employment for firms with 250 to 499 employees. Among the very largest firms, the share of total wages exceeds the share of total employment by almost 8 percentage points.

The contrast between the distributions of firms and employment by size class is striking, but it is important to consider it in context. The data presented here simply reveal that smaller firms have fewer employees and larger firms have more - which reflects exactly how the size classes have been defined. If a firm with four employees in 2010 experienced a net growth of even a single job over the subsequent year, it would move to the five to nine size class for 2011, taking all five jobs with it. This illustrates the importance of careful attention to definitions when analyzing size of firm data. Like a snapshot, the figures in Table 1 tell us how firms, employment, and wages are distributed at a specific point in time. To answer questions about dynamic processes such as job creation, we would need to examine data that was defined to measure change over time.

Oregon size of firm distributions tend to remain stable from one year to the next. The share of firms with one to four employees increased by 3 percentage points between 2001 and 2010, while the shares for all other size classes decreased by 1 percentage point or less (Table 2). Firms with fewer than 20 employees collectively increased their share of total jobs by almost 2 percentage points, with a proportional gain in their share of wages. Shares of employment and wages in most other size classes decreased slightly. Among firms with 500 or more employees, the share of total jobs increased a fraction of 1 percentage point, while the share of wages increased by more than 1 percentage point.

The U.S. experienced two recessions in the past decade, the first in 2001 and the most recent from 2007 to 2009. The distribution of firms and employees by size class did not change greatly during these two recessions in either Oregon or the nation as a whole. The share of firms in the smallest size class increased slightly during recessionary periods, as did the share of employees in the largest size class.

These observations are consistent with the figures presented in Table 2, and are likely connected with widespread layoff activity. As firms of every size cut jobs, some of them move into smaller size classes. Firms that retain at least 500 employees remain in the top size class no matter how many workers they lay off, which may account for the slight increase in the share of jobs for this size class.

The distribution of firms by size class in the first quarter of 2010 is quite similar for Oregon and the U.S. The largest difference is in the smallest size class, where Oregon's share of private firms is 3 percentage points higher than that of the nation as a whole. For every other size class, Oregon's shares are the same or slightly lower than shares at the national level (Graph 1).

Differences between Oregon data and national data are more pronounced when comparing employment distributions. In all but the largest size class, Oregon's share of employment is at least 2 percentage points higher than the corresponding U.S. share. As a result of this shift, Oregon's share of employment in the largest size class is more than 19 percentage points lower than the national share (Graph 2).

These differences are a natural consequence of the way firms are counted. A firm that operates in multiple states will be counted as one firm at the national level and classified based on its total number of employees in all states. For example, imagine a firm that has regional offices in all 50 states, with each regional office employing 10 people. At the national level this firm will be in the 500+ size class, while individual states will each count it in the 10 to 19 size class. All 500 of the firm's employees will appear in the 500+ size class at the national level, while each state will count only 10 employees. It is far more common for a firm to have 500+ employees throughout the entire United States than it is for a firm to have 500+ employees in a single state.

Oregon size of firm distributions tend to remain stable from one year to the next and are roughly comparable at state and national levels. Variations emerge, however, when comparing data for different industry groups.

Since firms with 250 employees or more accounted for about 35 percent of Oregon workers in 2010, it is worth asking what industries these larger firms belong to. Seventeen percent operate in the health care and social assistance sector, while another 17 percent are in manufacturing. Retail trade is the third-largest group, representing 15 percent of firms with 250 employees or more. Larger firms in the above three sectors employed between 46 and 53 percent of workers in their respective sectors in 2010 (Graph 3).

At the other end of the size-class spectrum, recall that 89 percent of Oregon firms had fewer than 20 employees in 2010 and accounted for about 27 percent of private-sector employment. These smaller firms employ a much larger share of workers within certain sectors. For example, 50 percent of employees in the professional and technical services industry worked in firms with fewer than 20 employees. In Oregon's construction industry these smaller firms accounted for 53 percent of that sector's employees in March 2010.

This article has used QCEW data to develop a snapshot of Oregon private-sector businesses by size of firm in 2010. Along the way we have explored measurement issues and definitions that help fashion the portrait that emerges. Understanding these details helps us to formulate questions our data can answer, and to perceive the limits of the answers we receive. For example, we know that 52,423 Oregon firms reported between one and four employees on their payroll in March 2010. But how many of these firms were "mom and pop" family businesses? How many were franchises? How many were local outposts of national or multinational enterprises? Even when we can't answer the question, we benefit from recognizing what we still don't know.