How much money do small businesses contribute to the economy? A practical, sourced explainer

What we mean by small businesses and why they matter

When analysts measure how small businesses help the economy they rely on standard program definitions and a few practical thresholds. The U.S. Small Business Administration and Census Bureau describe small firms in ways that depend on industry and employee cutoffs, and analysts usually work with those definitions when summarizing counts and shares.

Most public summaries emphasize that small firms are the large majority of U.S. business establishments, a core reason researchers focus on them in jobs and local-impact analyses. For a concise federal overview, see the SBA small business profile.

Public datasets commonly record three related items that matter for impact work: firm or establishment counts, employment and payroll, and the age of establishments. Statistics of U.S. Businesses records firm counts and employment by firm size and geography, while other sources capture payroll and business dynamics. For the Census Bureau overview of the program, see the SUSB program page.

Those three measures are different but complementary. Counts show how many establishments operate. Employment and payroll show how many workers and what wage base are tied to those establishments. Establishment age helps separate startups from older firms, a distinction that matters for net job creation.

Definitions used by SBA, Census and BLS

The SBA frames small business definitions partly by industry and partly by size, and Census SUSB provides establishment-level counts by employee ranges that researchers commonly use. Analysts must note that the same small-business label can mean different things depending on whether the study uses a 500-employee cutoff, a revenue test or industry-specific thresholds.

To review a federal profile with program-level definitions and national summaries, consult the SBA small business profile.

High-level snapshot of firm counts, employment and roles in the private sector

Short summaries from federal programs show that small firms account for nearly all U.S. establishments, which is why they are central to many local economic narratives. That pattern appears repeatedly in SBA and Census summaries.

When you read headline summaries about firm prevalence, remember those figures reflect establishment counts more than payroll or value added, which are measured separately and require mapping to national accounts for GDP comparisons.

How researchers measure the contribution of small firms

Researchers follow a two-step logic to estimate how small businesses help the economy: first they measure where workers and payroll sit across firm-size categories, and then they map those shares into national value-added or regional impact frameworks. The microdata supply the distribution, and national accounts convert distributions into GDP or value-added terms.

Key public microdata sources include SUSB employment and payroll by firm size and BLS business dynamics tables for flows of jobs. For primary data summaries and program access, see the Census SUSB landing page.

To convert microdata into a GDP-like share, analysts typically allocate BEA industry value added proportionally to firm-size shares observed in SUSB for the same NAICS cell. That proportional allocation is a pragmatic approach used in applied studies, but it depends on the assumption that payroll or employment shares reasonably proxy value added within each industry; for BEA industry data and accounts see the BEA industry pages.

The final step for many local studies is to apply input-output multipliers, such as BEA regional multipliers or RIMS II, to capture indirect and induced effects. Those multipliers translate direct payroll or value-added into total output or employment impacts in a region, and using the appropriate regional tool is standard practice.

Microdata sources and what they capture

SUSB provides establishment counts, employment and payroll by firm-size bins and by geography. These microdata are the starting point for many small-business share calculations because they directly record where workers and wages are located across firm sizes.

For workflows that track gross job flows and churn, researchers turn to BLS Business Employment Dynamics, which reports hires, separations and net changes across job cohorts and firm categories; see the BLS BED resource for those dynamics.

Linking microdata to national accounts and multiplier approaches

Linking SUSB microdata to BEA industry value added is usually an allocation exercise: map SUSB employment or payroll by NAICS to the BEA industry total, then allocate the industry value added by the observed proportion of employment or payroll in small firms. The method is clear but sensitive to assumptions about whether employment or payroll best approximates value added in each industry.

When analysts want to estimate total regional impact they then apply regional multipliers to the allocated direct value added. Analysts rely on BEA or RIMS II multipliers for that step and document the multiplier choice because different multipliers produce different total impact estimates.

Small firms and job creation: what the data show

Small businesses collectively employ a substantial share of private-sector workers, and they are an important source of gross job flows. Analysts using SUSB and BLS dynamics routinely emphasize the role of small firms in hiring and separations when describing private-sector churn; for BLS job-flow detail see the Business Employment Dynamics resource.

Gross job creation and destruction are an important frame because they show turnover rather than just net change; BLS BED measures that turnover across firm sizes and is the standard public source for gross flows.

Analysts also note that firm age matters: startups and younger firms tend to create a disproportionate share of net new jobs even though they represent a smaller slice of total payroll. International and sector studies underscore the importance of firm age for understanding job creation patterns; for international context see the OECD SME and Entrepreneurship Outlook.

When discussing job creation it is useful to separate gross flows, which capture churn, from net job growth, which often depends on a small subset of high-growth young firms rather than the entire small-firm population.

Share of private-sector employment and role in gross job flows

National microdata show small firms are a major presence in private employment and that they account for a material share of hires and separations observed each year. Use the Census SUSB tables to examine employment shares by firm-size class for a chosen geography and year.

To check patterns of hires and separations, consult BLS BED tables which break down gross job flows by firm characteristics and provide the time series researchers need to validate turnover hypotheses.

The importance of firm age and startups for net new jobs

Research from international organizations and entrepreneurship indicators highlights that startup activity and the youngest firms often drive net new employment even when older small firms remain important employers. For startup activity measures and trends consult the Kauffman startup indicators.

Because of that age effect, analysts who are interested specifically in net job creation often combine size-based analysis with age-based filters to isolate the contributions of new and young firms from the broader small-firm population.

Estimating small-business shares of GDP and value added

Estimating the GDP contribution of small businesses typically combines SUSB employment or payroll shares by firm size with BEA industry value added. Analysts map NAICS cells across the two datasets and allocate BEA value added proportionally to the small-firm share observed in SUSB for each industry.

Those proportional allocation steps are common in applied studies because BEA does not publish firm-size value-added breakdowns, so researchers use SUSB microdata as the best available proxy and rely on BEA totals for the industry-level value added that sums to GDP; see BEA industry data for the relevant accounts.

Applied studies that use this linkage find a wide range of reported small-firm GDP shares depending on choices about the allocation base, firm-size cutoffs and handling of owner compensation. Some methods emphasize payroll shares, others use employment, and those choices move the resulting percentage point estimates.

Another methodological complication is pass-through income and owner compensation, which can be recorded differently in employer payroll records versus national accounts. When analysts reconcile microdata with BEA accounts they must document how they treat owner pay and pass-through returns because that affects the estimated share of value added attributable to small firms.

Common calculation steps and why results vary

A straightforward calculation sequence looks like this: extract SUSB employment or payroll by firm size and NAICS for the target year and geography, obtain BEA industry value added for the same industry-year, allocate the industry value added to firm-size bands proportional to the SUSB shares, and sum across industries to get a national or regional small-firm value-added estimate.

Variation arises because analysts must choose whether to use payroll or employment as the allocation basis, which firm-size cutoff defines small firms, and whether to include adjustments for owner compensation or pass-through entities. Each choice moves the headline share up or down and should be reported as an assumption rather than presented as a single immutable fact.

Ranges reported in applied studies and sources of variation

Because of methodological differences, published ranges for the share of value added attributable to small firms vary from low single digits to mid-40s in percentage terms in different studies. That variation reflects definitional choices and allocation methods rather than a single data error, so readers should interpret point estimates as method-dependent ranges.

When presenting results, it is best practice to report a central estimate with a plausible range and a short note describing the primary assumptions, such as the allocation basis and handling of owner compensation.

Step-by-step approach for estimating local or sector impacts

This practical checklist shows the data and steps to estimate county or sector small-business impacts using public sources. Start by choosing the target geography and year, and then gather matching SUSB and BEA files for that year to avoid timing mismatches.

Required inputs commonly include county or NAICS SUSB employment and payroll by firm size, BEA industry value added for the same year, and a regional multiplier tool such as BEA regional multipliers or RIMS II to translate direct value added into total output or employment impacts.

After you prepare the inputs, map SUSB employment or payroll shares to BEA industry totals by NAICS, allocate value added proportionally to small-firm shares, and then apply a regional multiplier to estimate indirect and induced effects. For BEA industry totals and multiplier guidance see the BEA industry accounts pages.

Validate your headline job estimates with BLS job-dynamics tables to check whether recent gross flows or net trends are consistent with the direct payroll and multiplier results. BLS BED provides the hires and separations tables that are useful for cross-checking net job outcomes.

Data sources to download and how to align them

As a minimum gather SUSB employment by firm size and payroll by firm size, BEA industry value added for the same industries and year, and a regional multiplier table appropriate to your geography. Keep the data years aligned and document any adjustments you make for timing differences.

Common alignment steps include re-basing industry codes if necessary, aggregating or disaggregating NAICS cells to match across SUSB and BEA, and converting payroll to value-added proxies if you choose payroll as your allocation basis.

Applying multipliers and validating with BLS job dynamics

Choose a regional multiplier that fits your scope. BEA regional multipliers are standard for many applied studies, while RIMS II is another commonly used source. Document which multiplier you used and why.

After applying multipliers, compare the implied net job changes to BLS BED figures to ensure your estimated impacts are not inconsistent with observed hires and separations. If the multiplier-adjusted estimate implies a large unexplained job change relative to BED, revisit assumptions about allocation or multiplier scale.

Common mistakes and methodological caveats to watch

One frequent mistake is inconsistent firm-size cutoffs across data inputs. If you use a 500-employee cutoff in SUSB but then apply a different definition when interpreting value added you will create incompatible measures. Always state the firm-size cutoff used and apply it consistently.

Another common error is improper treatment of owner compensation and pass-through income, which can inflate or understate small-firm shares when payroll is used as a proxy for value added. Make an explicit choice about owner pay and document it for readers.

Timing mismatches between microdata and BEA annual benchmarks can change reported shares. Microdata like SUSB and BLS BED are updated on their schedules and BEA revises industry accounts on a different timetable, so align years or run sensitivity checks across nearby years.

Practical tips to avoid these pitfalls include always documenting the microdata year, stating assumptions about owner compensation, and reporting ranges rather than single points to reflect methodological uncertainty.

Errors that inflate or understate small-business shares

Using payroll without adjusting for proprietor income or pass-through returns can overstate the value tied to small firms relative to BEA-style value added. Conversely, excluding owner compensation from the small-firm allocation can understate their contribution. Either direction is possible depending on the treatment chosen.

To reduce the risk of misleading results, present both payroll-based and employment-based allocations as a sensitivity analysis and explain which you prefer and why.

Data timing and harmonization issues

Because BEA benchmarks and Census microdata update on different timetables, analysts should verify that industry totals and microdata years match or run the calculation for adjacent years to test stability. Document any re-basing or concordances used to align NAICS codes.

When possible, publish the reproducible spreadsheet or script that shows the mapping and allocation steps so others can verify the choices and re-run the analysis with alternative assumptions.

Illustrative scenarios and short examples

This section presents short, labeled sketches that show how the method choice affects headline estimates at different scales. The goal is to illustrate how proportional allocation and multipliers shape results without asserting a single national number.

National sketch. A common applied path is to take SUSB employment shares by firm size by NAICS, map those shares to BEA industry value added, and sum the allocated value added across industries to obtain a national small-firm value-added range. The resulting range depends on whether employment or payroll is the allocation base and on the firm-size cutoff used; for BEA industry totals see the BEA industry data.

County sketch. At the county level a direct payroll allocation often understates total economic impact because it omits indirect supplier and consumer-spending effects. Applying a regional multiplier increases the total impact estimate and is therefore a typical step when moving from direct payroll to total county impact estimates.

Sector sketch. Services sectors and manufacturing differ in how payroll maps to value added and in their local supply chain intensities, so a given small-firm employment share can translate into a larger or smaller share of value added depending on the sectoral mix and supply-chain structure.

Each of these sketches is meant to show direction and sensitivity rather than to provide a single definitive national number; analysts should run the same workflow on their chosen data and publish assumptions and ranges.

National example: combining SUSB and BEA to get a range

A simple national exercise uses SUSB employment shares by NAICS and BEA industry value added to produce a method-dependent range. Because BEA does not publish firm-size splits directly, this allocation approach is the pragmatic standard researchers use for national and cross-industry estimates.

When reporting national ranges, label the allocation basis and the firm-size cutoff so readers can see how assumptions affect the result and can re-run the calculation if they prefer alternative choices.

County example: how multipliers change the headline effect

At the county scale, a common pattern is that direct payroll measures show the immediate wage and salary connection but do not capture spending linkages. Using BEA regional multipliers or RIMS II expands the direct effect to account for supplier and household spending feedback.

Document the multiplier used and consider presenting both the direct and the total multiplier-adjusted figures so local readers understand how much of the headline effect is direct versus indirect.

Sector example: a service industry versus manufacturing comparison

Service industries often have higher shares of small establishments by count but lower average payroll per establishment than some manufacturing industries. This means that a high small-establishment share in services may not translate into a proportionally high share of value added without allocation adjustments.

When comparing sectors, map NAICS carefully and consider presenting payroll and employment allocations side by side to illustrate how the two bases produce different small-firm value-added estimates.

How to interpret estimates for policy, reporting and civic use

Choosing a method depends on the audience and the question. For quick public reporting (see news), a payroll-based allocation with clear assumptions may be adequate. For policy decisions that will affect budgets or incentives, a more conservative approach with sensitivity analysis and external review is advisable.

When presenting results to a non-technical audience, frame findings as ranges with a short explanation of the main assumptions, and include a note on where to find the underlying public data so others can verify the work.

Decision criteria for choosing methods and reporting results

Decide upfront on the audience, geographic scope and acceptable assumptions. If local policymakers need conservative estimates, favor conservative multipliers and document uncertainty. If the aim is exploratory analysis, present a central estimate with sensitivity bounds.

When in doubt, state the assumptions explicitly and offer to provide the data mapping or spreadsheet for peer review or journalistic fact checking; contact us.

How readers should treat ranges and uncertainty

Ranges reflect methodological choices more than measurement error in most cases. Report ranges and explain the largest drivers of variation so readers understand why a different choice could produce a different result.

If a report will inform public discussion, recommend that journalists and local officials treat single-point estimates cautiously and look for reproducible inputs and documented assumptions before drawing policy conclusions.

Key takeaways and where to find the primary data

Small firms account for the large majority of U.S. establishments and are an important part of private employment and job flows, but translating those facts into a share of GDP requires careful mapping to BEA industry accounts. For a federal program summary see the SBA small business profile.

Primary data to consult include SUSB for establishment counts, BLS BED for gross job flows, and BEA industry accounts and multiplier tools for value-added and total-impact calculations. For the Census SUSB program see the SUSB pages and for BEA accounts see the BEA industry data pages.

For readers interested in civic contexts, noting the role of firm age and startups alongside firm size gives a fuller picture of how small firms contribute to net new jobs and local economic dynamism; OECD and Kauffman indicators provide complementary context on startups and entrepreneurship trends. (see about)