IRS Statistics of Income tables provide official tax-record data on income sources, including wage and salary income and self-employment income. These administrative records are the type of primary source commonly used to compare earnings across self-employed and wage-employed taxpayers.

In new research, Minneapolis Fed consultants Anmol Bhandari and Ellen McGrattan and coauthors used an extensive administrative dataset of IRS and Social Security records. Across the full sample of taxpayers between 2000 and 2015, workers classified as primarily self-employed had an average annual income of almost 60 percent more than paid-employed workers. By age 55, the average annual income for a worker who had tried self-employment was nearly twice as large as for a paid-employed worker.

This paper uses the 2004–2016 Health and Retirement Study linked to administrative earnings records and finds substantially more self-employment income in the survey than in the administrative data. It also reports that the estimated self-employment rate is 25.4 percent in the HRS compared with 15.1 percent in the DER, and notes that self-employment income is under-identified in both administrative earnings records and major household surveys.

Our results confirm prior research showing that firm size is positively associated with wages: workers at larger firms earn more, on average, than workers at smaller firms. We also find that firm size is positively associated not only with higher wages, but also with greater employment security during the COVID-19 crisis.

The study defines and measures ‘hard’ job quality (e.g., wages, working hours, physical environment) and ‘soft’ job quality (e.g., autonomy, relationships) for wage workers and examines their impact on health. It finds: “the level of the overall health of wage workers was increased as the hard and soft quality of jobs was increased… The standardized beta coefficients of hard job quality and soft job quality were 0.316 and 0.079, respectively, confirming that the effect of hard job quality was about 4 times stronger than the soft job quality on the overall health of workers.” Although this paper focuses on employees rather than entrepreneurs, it provides a framework and empirical evidence on wage and non‑wage job quality dimensions in paid employment that are often compared to conditions in entrepreneurial jobs.

The NBER working paper version of the study documents that new firms pay wages about 26 percent lower in the sample, but the apparent gap is largely explained by differences in worker quality and observable characteristics. The paper concludes that, for a given worker who has job opportunities at similar-quality firms, the expected wage penalty of working at a new firm is economically insignificant on average.

“Startups create good jobs. Startups pay starting salaries that exceed the area’s average wage. In 2021, startups paid an average of $73,190 in starting wages compared with the average wage of $65,525.” ... “In 2021, average wages in tech firms in Kansas City start 42% higher than the average wage ($100,690 vs. $65,525). The tech startup average wage, non-tech startup average wage, all startup average wage and all establishments average wage are compared in the graph below, along with the living wage and poverty wage.”

This 2025 study examines wages and broader job quality in B Corporations, many of which are young or mission‑driven entrepreneurial firms: “This research has focused on identifying, characterizing, and classifying relationships between elements of jobs and worker well-being, encompassing not only wages and benefits but also scheduling, advancement opportunities, and worker voice.” The authors report mixed findings: some certified B Corps offer above‑market wages and strong benefits, but on average “wage levels are not uniformly higher than those of comparable firms in the same sectors,” suggesting that mission‑oriented entrepreneurship does not automatically translate into higher average pay relative to traditional wage employment.

The paper compares labor-market outcomes across employment types and countries and provides context on the prevalence of self-employment versus wage employment. It states that own-account workers account for only about 5% of employment in the United States, while wage and salary workers account for about 85% of employment.

“Self-employed workers have significantly higher average incomes and steeper, more persistent income growth profiles than their paid-employed counterparts.” “By age 55, the self-employed earn an estimated average income of $134,000 compared to $79,000 for paid-employed workers with similar characteristics. This pattern holds across various subgroups defined by industry, skill level, and demographic attributes.”

Using linked payroll and tax records, this paper studies earnings differences between self-employed and wage-employed workers. The research emphasizes that self-employed earnings are highly dispersed and that the average can mask substantial inequality, with a small share of self-employed workers accounting for a disproportionate share of self-employment income.

We compare the earnings trajectories of individuals who take jobs at startups with those of similar workers who take jobs at established firms in the same local labor markets. Although startup employees initially earn slightly less than their peers at established firms, their earnings grow more quickly. By 10 years after the job change, startup employees, on average, earn more than comparable workers who joined established firms.

The paper examines the cross-sectional distribution of labor earnings for workers in paid employment and for the self-employed, using cross-country data. It is directly relevant to comparisons between self-employment and wage employment, focusing on how institutional settings and labor-market conditions affect relative pay.

Haltiwanger, Jarmin, and Miranda analyse US employer data to distinguish job creation by young and small firms from that by more established employers. They find that young firms (a proxy for entrepreneurial businesses) are responsible for a disproportionate share of gross job creation. However, in discussing job quality, they note that “employment at young and small firms tends to be associated with lower average wages and greater volatility than employment at more mature firms,” indicating that the jobs created by entrepreneurship are not generally higher‑wage than those in incumbent firms in the same labour market, even though they contribute significantly to net employment growth.

The abstract states: “This paper analyses the importance of entrepreneurs in terms of job creation and wage growth. Relying on unique data that cover all establishments, we investigate how much of aggregate job creation can be attributed to entrepreneurial firms and how wages develop in these firms compared with other firms.” The description indicates that the study evaluates both employment creation and wage dynamics in entrepreneurial firms relative to non-entrepreneurial firms in the same economy, but the abstract page does not give specific wage level comparisons in numeric terms.

Hamilton and co‑authors review empirical evidence on the earnings of self‑employed entrepreneurs relative to comparable wage workers. The paper concludes that “a large fraction of entrepreneurs earn less than they would in paid employment with similar characteristics,” and many accept this earnings discount in exchange for non‑pecuniary benefits such as autonomy. While the focus is on the earnings of entrepreneurs themselves rather than their employees, the findings challenge the notion that entrepreneurship is systematically associated with higher monetary rewards than wage employment in the same labour markets.

Average monthly earnings of employees in 2011 also favored the established firms. With over $3,200 per month, the only group to come in above the all-firm average of $3,100 was the group with companies 11 years or older. The newest companies had the lowest average monthly earnings with $2,116 – 55 percent less than the oldest firms.

The paper notes: “Therefore, unincorporated self-employment has both advantages and disadvantages compared to wage employment, facilitating the transition between these two states.” While the main focus is on the impact of minimum wages on nonemployer (self-employed) businesses, the discussion highlights that self-employment and wage employment differ in earnings risk and flexibility; however, the paper does not provide a general statement that self-employment earnings are always higher than those of wage employment in the same labor market.

For the typical worker, the smaller the employer, the lower the pay. The median worker at the smallest businesses, those with fewer than 20 employees, makes about $21,000 less than the median worker at companies with 500 or more employees. When we look at a matched sample of workers—people working at the same employer in May 2024 and May 2025—we see that not only is pay lower, median pay growth at small employers lags that at larger ones.

We document that business owners, particularly new entrepreneurs, tend to earn less on average than comparable wage and salary workers. Even conditional on survival, entrepreneurial earnings are highly volatile and typically lower than those of similar individuals who remain in paid employment.

This study assesses the quality of jobs created by private‑sector clients of development finance institutions, many of which are entrepreneurial firms in emerging markets. It explains that the survey “includes standardised questions, and allows for tailored questions, including job quality aspects such as participation, training, … wages and benefits.” The report finds substantial variation in wages and other job quality indicators across entrepreneurial firms and sectors; while some supported enterprises pay above local averages, others do not, indicating that jobs created through private entrepreneurial projects are not systematically higher‑wage than existing employment opportunities in the same labour markets.

The article notes: “Startups play an outsized role in net job creation in the United States, but they do not consistently offer higher pay or greater job stability than incumbent firms.” It further explains that, on average, “workers at young firms tend to earn less than those at older firms, although there is substantial variation and some high-growth startups pay exceptionally high wages.”

In labor economics, claims about whether entrepreneurship creates higher average wages usually require comparing like-for-like workers in the same labor market, not comparing founders’ income to employee wages. Raw averages at entrepreneurial firms are often affected by worker sorting, firm age, occupation mix, and founder compensation structure, so adjusted estimates are the relevant test.