Growthology

Self-Employment Earnings Puzzle

I have been reading "The Entrepreneurial Earnings Puzzle: Mismeasurement or Real?" a working paper by Thomas Åstebro and Jing Chen, that looks at the research literature on entrepreneurs' earnings (paper available here). See the paper for a literature review; research generally finds, among other things, that entrepreneurs earn on average less than employed workers and work longer hours, and continue to do so despite having better opportunities by being an employed worker.

With a supposed earnings disadvantage, why would anyone choose to be an entrepreneur willingly? Åstebro and Chen specifically investigate if the earnings gap between self-employed and employed workers is due to underreported income of the self-employed. They base their analysis on food consumption patterns—take two people with the same income, self-employed and employed workers, and see how their food consumption differ. They find that self-employed spend more on food than similar employed persons, suggesting an underreporting of income by the self-employed by significant degrees. This assumes that food preferences are the same across individuals (among other assumptions), so the authors conclude that this is partial, suggestive evidence of underreporting of income. Regardless, it does not account for other theories on why people choose to become entrepreneurs.

My primary takeaway about entrepreneurial earnings:

"In general, predictors of entrepreneurial earnings are weak... As researchers, we must remain humble regarding our ability to explain or predict entrepreneurial success. By far the most prominent explanation is chance."

What’s the Best Way to Hire Employees?

New research by Stephen V. Burks, Bo Cowgill, Mitchell Hoffman, and Michael Gene Housman (paper available here) looks at the performance of employees hired as a result of a referral vs. those hired from other sources. They knew from previous research that referred employees often performed better, but couldn't explain why.

Through their research, they found that, relative to non-referred workers, referred workers are a better deal, even though both groups have similar characteristics and score similarly on standard productivity measures. Referred workers:

• Are less likely to quit
• Are more innovative
• Have fewer accidents

The graphs below shows worker likelihood of quitting based on referral status:
Referrals result in employees who are better suited for the job at hand, which, their research found, makes them 25 percent more profitable than non-referred workers. (See the math in Table 16 below.) Also of note, is that "people tend to refer others like themselves, making referrals from high productivity workers especially valuable."



The practical take-away here for entrepreneurs? If you can find employees through referrals from trusted sources, do so. If you don’t have one already, start an official employee referral program to find the best-fit candidates. If you already have such a program, you’re good to go for now. The authors are planning future research to explore the benefits of expanding existing employee referral programs.

Immobility and Freelancing

I'm trying to square two seemingly contradictory trends. First, as we've known for several years, geographic mobility in the United States is today half of what it once was. This applies to interstate, intrastate, and intra-county moves. At the same time, apparently, we are living in a new era of freelancing and independent work, such that 58 percent of Millennials surveyed called themselves an entrepreneur because of freelancing work. That Forbes article also references an Intuit study purporting to find that 40 percent of US workers will be freelancers in 2020, only seven years from now. That seems astoundingly high, and would seem contrary to the steady downward trends in mobility, which predate the recession.

So what's going on here? One possibility is that the data are wrong. It seems unlikely that decades of Census data, based on IRS records, are incorrect. Self-reported survey responses seem a likelier candidate for error. It's important to note, too, that the Intuit report identifies "contingent employees" and contractors as the source of this freelancing growth, which is not necessarily the same as what we typically associate with freelancers. The latter term has a certain romance about it and conveys a sense of independence and control which, not surprisingly, are the attractions identified in the Millennial survey.

Identifying oneself as a "contingent employee" doesn't really have the same ring to it and, in fact, probably highlights some economic forces at work here. I was speaking to someone the other day whose twenty-something daughter is a freelance illustrator who wants nothing more than a steady job. Companies utilizing the services of freelance illustrators, according to this one source at least, solicit samples based on a proposed project--with the condition that the company owns the resulting intellectual property. What inevitably happens is that scores of freelancers submit bids, one is chosen, and the company is free to use the other "free" bids.

Contingent employees and contractors are not owed benefits and so are perhaps a more attractive option for companies in certain sectors. A few years ago, the IRS made noise about cracking down on companies' classification of workers as freelancers and contractors, but I'm not sure where that ended up. In any case, we should take care when celebrating the increase in freelancing because it's not always the romantic notion we assume it be. And it's notable that such large percentages of people in the aforementioned survey said they wanted to leave their job to be independent--no word on the subsequent revealed preferences and resulting satisfaction. Self-employment rates have not risen alongside the reported increases in freelancing, with the exception of a one percentage point increase in the incorporated self-employed.

(Worth another blog post: lots of people seem to equate freelancing with being an entrepreneur. That's one of the points of the Forbes article, that "being an entrepreneur" today is more about freelance work than starting and building a company. That's highly debatable.)

But the larger puzzle remains: if freelancing is really on the rise and set to rise further, why has geographic mobility been falling, and will that suppress the expected increase in freelancing?

It could be the case that people reporting themselves to be freelancers still hold a steady day job and engage in supplemental work through places like Etsy. So perhaps we are missing large chunks of economic activity. Perhaps a rise in freelancing actually explains falling mobility? If you are a freelancer/contractor/contingent employee, depending on the sector, you need to be in a location with a dense concentration of relevant companies. The illustrator mentioned above would probably have more difficulty in Helena, Montana, than New York City. This presumes that the physical density of some industries has been rising; I don't know to what extent that is true.

Alternately, the opposite could be true: perhaps the severing of the relationship between physical location and work means these footloose workers choose a place to live based on quality of life because all they need is an Internet connection to manage their freelance/contingent career. That would also presuppose a narrowing of the places that are attractive to freelancers. In its last State of Metropolitan America report, the Brookings Institution pointed out that while the country as a whole is aging, certain metro areas are actually "younging" because they have become magnets for twenty- and thirty-somethings. If it's true that Millennials are more predisposed to the contingent career than older Americans, this could mean a close link between freelancing and immobility. (At the same time, however, self-employment rates are generally much higher among older age groups.)

So I don't know what to think. Geographic mobility is undoubtedly down; we know that. Perhaps this "boom" in freelancing is a combination of self-reported classifications and people engaging in more freelance activity on the side, while they continue to hold down a job. And I suppose that with more companies using contingent labor--perhaps facilitated by the Affordable Care Act's exchanges--these two seemingly contradictory trends could continue to coexist. Or maybe all freelancers and contractors and contingent workers will simply move to the states that set up those health exchanges.

Signaling and Investor Preferences

I mentioned last week a paper by Jing Chen on serial small business owners. From the same special issue of the Journal of Economics & Management Strategy is another paper, "Show Me the Right Stuff: Signals for High-Tech Startups" by Annamaria Conti, Marie Thursby, and Frank T. Rothaermel (available online here [gated]).

The authors construct a theoretical model to examine the influence that (i) obtaining patents and (ii) receiving financing from founders, friends, and family (FFF) have on follow-on investments by angel capitalists and venture capitalists in high-tech startups. They then analyze data from 226 startups that went through the Advanced Technology Development Center at Georgia Tech (ATDC).

Conceptually speaking, patents and FFF investments serve as signals of quality to outside investors: willing to invest time and money in procuring a patent; willing to commit your own money and the money of close connections. What Conti, Thursby, and Rothaermal's analysis of the ATDC startups suggests is that angel investors and venture capitalists do not value these quality signals the same way: VCs put more weight on patents, while angels put more weight on FFF funding. That is, VCs appear to care more about the underlying quality of the startup's technology while angels care more about the commitment of the founders to the startup.

The implication for founders is that you should consider which signal to invest in depending on who you want to target for follow on funding; or if you have already invested in one versus the other, know which audience to target going forward. Both patents and FFF funding represent different costs and appear to have significantly different values to different types of investors.

Clusters of Entrepreneurship and Innovation

The title is from a new paper by Aaron Chatterji, Edward Glaeser, William Kerr, available here, about local entrepreneurship (disclosure: the Kauffman Foundation supported the conference at which this paper was presented). When I talk to entrepreneurs about policy, they comment that they wish they knew more about what to do at the local level. Successful entrepreneurs will frequently be tapped by local policy makers for their input or to give a speech, and they sometimes feel overwhelmed or at a loss for words when it comes to talking about local government policy.

This paper reviews the literature on this subject and plainly stated it is excellent. I recommend it to a wider audience because it is not only good, but written in plain english, unlike many academic papers. I will read this multiple times, as it serves to refresh my memory on some papers and inform me about new literature I'm not familiar with yet.

Hat tip to Robert Strom for pointing this paper out.

Serial Small Business Ownership

Serial entrepreneurs generally do better with their subsequent businesses than first-time entrepreneurs do with their first business.This has led to the belief that serial entrepreneurs' previous business experience sets them up for greater success with their subsequent ventures. However, it is also possible that serial entrepreneurs simply have better ability at the start.

A forthcoming paper, "Selection and Serial Entrepreneurs" by Jing Chen (available online here [gated]), looks at this issue among small business owners and the self-employed and finds that natural ability of serial business owners is the dominant factor for why they do they better; that is, their success is independent of previous venture experience.

Chen looks at self-employment and serial business ownership data from the National Longitudinal Survey of Youth 1979 dataset from 1981 to 2008. There are a lot of caveats about the data used, which focus on self-employment and small businesses, so to the extent findings extrapolate to growth-oriented startups is debatable. For example, 43 percent of the serial entrepreneurs changed industries with their subsequent businesses. So when Chen finds that entrepreneurial experience does contribute to improved outcomes when those entrepreneurs start a subsequent business in a familiar industry (rather than changing industries), it casts some doubt in my mind about the degree to which experience is or isn't influential. Regardless, the study highlights the role of innate entrepreneurial ability, specifically for small business owners and the self-employed--natural ability appears to have a dominant effect on firm performance.That is, the generalized implication is that innate ability perhaps matters more than experience.

Disclosure: The Kauffman Foundation is supporting a data module being added to the National Longitudinal Survey of Youth 1979 (NLSY79), the author's source data for this study, and the special issue of the journal this paper is published.

Best Worst Startup Ideas

I have tended to stay away from reposting from other blogs so I don't clog up people's feeds with spam, but Tyler Cowen has pointed out something irresistible.

From Quora, the most ridiculous startup ideas that have succeeded:

• Facebook - the world needs yet another Myspace or Friendster except several years late. We'll only open it up to a few thousand overworked, anti-social, Ivy Leaguers. Everyone else will then join since Harvard students are so cool.
• Dropbox - we are going to build a file sharing and syncing solution when the market has a dozen of them that no one uses, supported by big companies like Microsoft. It will only do one thing well, and you'll have to move all of your content to use it.
• Amazon - we'll sell books online, even though users are still scared to use credit cards on the web. Their shipping costs will eat up any money they save. They'll do it for the convenience, even though they have to wait a week for the book.
• Virgin Atlantic - airlines are cool. Let's start one. How hard could it be? We'll differentiate with a funny safety video and by not being a**holes.
• Mint - give us all of your bank, brokerage, and credit card information. We'll give it back to you with nice fonts. To make you feel richer, we'll make them green.
• Palantir - we'll build arcane analytics software, put the company in California, hire a bunch of new college grad engineers, many of them immigrants, hire no sales reps, and close giant deals with D.C.-based defense and intelligence agencies!
• Craigslist - it will be ugly. It will be free. Except for the hookers.
• iOS - a brand new operating system that doesn't run a single one of the millions of applications that have been developed for Mac OS, Windows, or Linux. Only Apple can build apps for it. It won't have cut and paste.
• Google - we are building the world's 20th search engine at a time when most of the others have been abandoned as being commoditized money losers. We'll strip out all of the ad-supported news and portal features so you won't be distracted from using the free search stuff.
• Github - software engineers will pay monthly fees for the rest of their lives in order to create free software out of other free software!
• PayPal - people will use their insecure AOL and Yahoo email addresses to pay each other real money, backed by a non-bank with a cute name run by 20-somethings.
• Paperless Post - we are like Evite, except you pay us. All of your friends will know that you are an idiot.
• Instagram - filters! That's right, we got filters!
• LinkedIn - how about a professional social network, aimed at busy 30- and 40-somethings. They will use it once every 5 years when they go job searching.
• Tesla - instead of just building batteries and selling them to Detroit, we are going to build our own cars from scratch plus own the distribution network. During a recession and a cleantech backlash.
• SpaceX - if NASA can do it, so can we! It ain't rocket science.
• Firefox - we are going to build a better web browser, even though 90% of the world's computers already have a free one built in. One guy will do most of the work.
• Twitter - it is like email, SMS, or RSS. Except it does a lot less. It will be used mostly by geeks at first, followed by Britney Spears and Charlie Sheen.

Is starting a business in your home area a good thing?

The Kauffman Foundation is interested in how geography affects entrepreneurs these days (here, here, and here), so when I came across a paper about entrepreneurs’ location choice, it piqued my interest.

"Home sweet home: Entrepreneurs' location choices and the performance of their ventures," a fascinating paper by Michael S. Dahl and Olav Sorenson (available here), looks at Danish data to determine if where entrepreneurs choose to start their businesses affects how well the new business performs.

Entrepreneurs, more so than employees, tend to live in the area of their birth, or to start companies in places they have lived for a long time. As we might expect, Dahl and Sorenson discover that "Ventures perform better – survive longer and generate greater annual profits and cash flows – when located in regions in which their founders have lived longer." Logically, as they know the area well and have numerous connections that help their new businesses, most importantly with raising capital and staffing the venture.

There were a few additional findings also of interest that arose from the examined data. Companies tended to perform better when founders were:

• Male
• Married
• Had higher levels of education
• Had wealthier parents
• Around the peak age of 42

Founders’ prior experience in the industry also adds to the odds of greater success for the business, especially if they are already familiar with the region as well. The authors note:

That fact raises interesting implications for the understanding of agglomeration. In general, the literature on agglomeration has assumed that firms must cluster to benefit from collocation. Our results, however, suggest that rational entrepreneurs would enter the industries in which they have employment experience and locate their businesses in the regions in which they have lived. Since acquiring experience at some existing firm in the industry generally implies living in close proximity to it, the implication of these constraints is that entrants in an industry will disproportionately emerge in regions that already have dense populations of firms in the same industry even if co-location itself offers no advantages [author emphasis] (Sorenson and Audia 2000).

So yes, start your business wherever "home" is for you. Being well-educated and having prior experience in your company’s industry would also help.

Search Engine Marketing - Worth It?

I am reading a very interesting paper by Thomas Blake, Chris Nosko, and Steven Tadelis about eBay and search engine marketing. They find through an experimental design that there is virtually no significant effect on eBay sales when they stop paid keyword search placements. One of the critical factors was using Google’s ability to tweak keywords for different geographies. They find that while new or infrequent users are drawn in by the search ads, their sales are minimal compared to the swath of repeat customers who don't need the ads to direct them to eBay. There is a discussion of implications about the (lack of) value of search engine marketing for large, established companies in the paper.

But startups are far from your typical large, established companies. I’m not sure how effective search engine marketing is for startups, and this paper doesn’t speak to that audience. What this paper does offer is some general evidence about organic versus paid search traffic, which is relevant to startups, particularly if one becomes more widely known. And it critically cautions against the pitfalls of a straightforward return on investment calculation (bold is my emphasis):

Not only do most consulting firms who provide marketing analytics services use observational data, recommendations from Google offer analytical advice that is not consistent with true causal estimates of ad effectiveness. As an example, consider the advice that Google offers its customers to calculate ROI:

“Determining your AdWords ROI can be a very straightforward process if your

business goal is web-based sales. You’ll already have the advertising costs for

a specific time period for your AdWords account in the statistics from your

Campaigns tab. The net profit for your business can then be calculated based

on your company’s revenue from sales made via your AdWords advertising,

minus the cost of your advertising. Divide your net profit by the advertising

costs to get your AdWords ROI for that time period.” [a footnote cites this quote to here].

This advice is very much akin to running a regression of sales on adwords [sic] expenditures, which is not too different from the approach that result in the inflated ROIs that we report in columns 1 and 2 of Table 5. It does not, however account for the endogeneity concern that our study highlights where consumers who use paid search advertising on their way to a purchase may have completed that purchase even without the paid search ads.

There’s nothing stopping from you conducting your own experiment if you wanted to go down that route.

This Week: Economic Bloggers Forum 2013

On this Friday, April 11, Brad DeLong will be hosting the 5th Annual Economics Bloggers Forum here at the Kauffman Foundation. The event will be livestreamed here: http://new.livestream.com/kauffmanfoundation/ebf2013. Last year, audience members and sometimes even panelists while they were up on stage were live tweeting their activities (hashtag TBD). I'll be working on a post-event report I hope to draft speedily. DeLong has a draft of his introductory remarks on his blog (I have been to many conferences that run over time so I particularly enjoyed this note: "Moderators are encouraged to, in the words of King Reheboam, chastise those who fail to respect the discourse not with whips, but with scorpions.")

Lessons for Entrepreneurs from Crowdfunding Redux

A couple of weeks ago, Mindee Forman highlighted some key findings from Ethan Mollick's research on crowdfunding. I'm seeing more and more papers crop up on this topic, including what I will be highlighting today, "Crowdfunding Creative Ideas: The Dynamics of Project Backers in Kickstarter," a working paper from Venkat Kuppuswamy and Barry L. Bayus (available here). The paper is interesting throughout and has some factoids about Kickstarter projects that I will not call attention to here.

This paper corroborates many of the findings that Mindee wrote about--that being featured on Kickstarter, having a video, and having a shorter project duration increases the chances of having a successful fund raise. While this may seem redudant, it's important in terms of research for different analysis to draw the same conclusions. We also learn some new information from Kuppuswamy and Bayus about attracting backers:

1. Smaller goals are more successful at being funded.
2. Many reward categories increases the chances for being funded.
3. A project with a colon in the title is correlated with successful projects [amusing, I wish they spent more time theorizing on this one].
4. Prior Kickstarter experience helps a little.

They also investigate donations along the funding cycle. They find that after initial backers sign on, an average project experiences a lull in attracting new backers, but makes a comeback once the deadline nears.

The authors interpret this in terms of the buzz or excitement a project intially generates, followed by a decrease in interest, followed by a pickup when the deadline approaches. Alternatively, to me this sounds like it could be a good samaritan dilemma--I see that someone else is already helping, so I don't need to contribute here; oh wait, this is more urgent than I thought, I should help!

In any event, Kuppuswamy and Bayus find that posting private and public updates as the funding deadline approaches helps quite a bit, which is significant because this is something under the project creator's control.

Lessons for Entrepreneurs from Crowdfunding

There is a lot of buzz out there these days about crowdfunding. As Kauffman Dissertation Fellow Ethan Mollick at the University of Pennsylvania’s Wharton School puts it, "Crowdfunding has been drawing substantial attention from policy makers, managers, and entrepreneurs, but relatively little notice from academics, even though it touches on many topics of importance to scholars of entrepreneurship."

His paper from last July, "The Dynamics of Crowdfunding: Determinants of Success and Failure" (available here), examined almost 47,000 projects on Kickstarter that raised a combined revenue of $198 million. Mollick concluded that several factors influence whether a project will succeed or not:

1. The greater the size of the founder's social network, the greater the chance for success (particularly Facebook in this case; this is also known as the 'be popular' strategy).
2. The underlying quality of the projects – those with high-quality, polished pitches are more likely to be funded (e.g., use a video; as Kickstarter's website states, "Projects with videos succeed at a much higher rate than those without.").
3. A strong geographic component tie-in seems to increase success (pitching country music in Nashville, film in Los Angeles, etc.).
4. A shorter Kickstarter duration is better (35 percent chance of success for 30-day pitches, 29 percent for 60-day pitches). Mollick noted that a longer duration implies a lack of confidence in the project’s success.
5. Being highlighted on the Kickstarter website is hugely beneficial (89 percent chance of success vs. 30 percent for unfeatured projects).
6. A large number of creative individuals in the city where the project is based is associated with greater success (target these kinds of people).

Geographic Distribution of Projects by Success

The circles on this map represent counts of Kickstarter projects by city; the larger the circle, the more projects based in that area. The shading within the circle reflects the portion that were successful—dark green represents successfully funded projects while light green indicates the project was not funded. Based on Mollick's research, odds are that the successfully funded projects in given cities were a good fit culture-wise for that city.

The Kauffman Foundation will be writing some papers on crowdfunding in the near future – stay tuned!

Preliminary Look at 1 Million Cups

Yasuyuki Motoyama and I have a paper out today about 1 Million Cups (1MC), a program of Kauffman Labs for Enterprise Creation. 1MC brings together entrepreneurs weekly on Wednesday mornings--two startups talk about their businesses for about six minutes each, followed by Q&A for twenty minutes with the audience, which is compromised of anyone who wants to attend, typically other entrepreneurs, mentors, advisers, and other supporters of entrepreneurship. The program is free and as the name suggests provides ample coffee to attendees. We conducted a survey of attendees of the program back in November 2012, or roughly seven months after the program first launched in Kansas City.

There is a press release about the survey findings from Kauffman here; the paper is here; we have an article in Huffington Post about some high-level findings going up later today and I will update this post when it goes live. **Update: Huffington Post article is here. Last but not least, Jordan Bell-Masterson put together an interactive chart about some of the survey data about other Kansas City programs here.

I think the paper is particularly relevant for community organizers who are starting new programs, especially in small to medium-sized regions. I'm tempted to write more but it would just be regurgitating what is written elsewhere, so I'll just say check out the paper if you're interested in learning more.

Venture Capitalists Prefer Vegetables

Very quickly before I start, I want to point out this research would conceivably extend to all investors that take a sizable chunk of preferred stock in a startup, not just venture capitalist (my opinion, not stated by the authors).

VCs have to work with founding teams and other common stock shareholders when they want to exit their investment, and as initial public offerings have decreased, trade sales, which were already common occurrence, have become even more prevalent in recent years. In a new paper, "Carrots & Sticks: How VCs Induce Entrepreneurial Teams to Sell Startups," (available here) Brian J. Broughman and Jesse M. Fried analyze 50 trade sales of VC-backed startups located in Silicon Valley.

Broughman and Fried investigate how VCs choose to persuade the entrepreneurial teams and common stockholders to make the sale—either through bonuses or carve-outs to common shareholders (carrots), or through force by terminating the entrepreneurial team, professionally blacklisting the founders, or trying to manipulate shareholder voting (sticks).  They find that VCs rarely overtly use sticks but do use carrots with some frequency—at least one carrot was provided to the entrepreneurial team in 45 percent of sales, and on average amounted to 9 percent of the deal value.

The practical application here is that you can hold out on a sale. The infrequent overt use of sticks suggests that VCs don’t want to hurt their reputation by being forceful. There are a couple of other discussion points in the paper you can check out for yourself. I give credit to the authors for writing in an understandable and non-technical manner.

Hat tip to Robert Strom for pointing out this paper.

LegalZoom and Kauffman Annual Survey

In 2012, the Kauffman Foundation partnered with LegalZoom to conduct a survey series of new business owners in an effort to establish a broader range of metrics than are available in more traditional data sets. Q1-Q4 saw confidence surveys that helped shine light on entrepreneurs’ particular take on consumer confidence and overall economic health.

At year’s end, we followed up on these confidence indexes with a final survey of 1,431 newly incorporated businesses focused on identifying founder and venture characteristics. The official report already offers a good summary of the important details gleaned from the data, so let’s instead explore some of the data’s limitations and what they mean for this survey and entrepreneurship statistics beyond it.

The scope of the survey starts coming into focus with a breakdown of the financing. When respondents were asked where they obtained startup capital, personal funding dominated all other external sources (83% of ventures used personal funding; 59% used it as the sole source). Investors backed just 7% of ventures in the sample, and only 5% of owners secured bank loans. With most funding coming out-of-pocket, we should anticipate these businesses to be relatively small, and the data back up this expectation. 82% of companies in the sample had revenues between 0 and $49k, and 70% had no employees other than the owner(s).

Given our sample’s slender financial profile, it is unsurprising that the top business type among respondents was “Consulting” (followed by “Other” and “Service: Other Services”).  First, we should account for the possibility of selection bias here – self-employed consultants are likely to have more time and therefore greater inclination to answer a survey than do owners with responsibilities to multiple employees and complex revenue streams. Nonetheless, we should not expect selection bias to account for the entirety of such a heavy weighting toward this type of business activity.

Accordingly, we should be wary not only of our own data’s limitations, but also of any statistics on small business which do not specify revenue or employment. For example, should a city or state boast responsibility for having “x” new businesses started in 2012, it’s worth the deeper dive to see how many of those will truly be impact businesses that bring jobs and money to the region. The research of Davis et al. demonstrates that nonemployers can and do migrate to the employer-universe, becoming a substantial source of revenue and growth in the period surrounding the transition.  Nonetheless, only 3% of nonemployers ever make this transition, meaning that our sample’s economic footprint is bound to be very limited given the size of the survey.  Nonemployers of all stripes play a role in the economy, but policymakers will want to focus primarily on those with growth aspirations, making “new businesses” with no further distinction a crude, weak metric.

Finally, perhaps more interesting than any information contained in the survey was the information that couldn’t be included. Among the respondents, 91% worked on their business for more than a month before incorporating. 60% worked on it for more than 6 months, and 35% did so for over a year. This is the black box of entrepreneurial studies, and the biggest problem facing the field: how do we obtain information about businesses in that crucial planning and preparation period? Could we only obtain information about that window of time with retrospective surveys, and if so would the data be tainted by the imperfections of memory? If we were able to contact entrepreneurs as they toiled through the development phase, would the very act of observation change their actions and invalidate the experiment? We may be no closer to answering these questions today, but our respondents have soundly reaffirmed the need to try.

Entrepreneurs Matter to Firm Growth

How much do individual entrepreneurs really matter to their firm’s performance? We intuitively suspect and anecdotally hear about how hugely important the individual entrepreneur is to a firm. A working paper by Sascha O. Becker and Hans K. Hvide, “Do Entrepreneurs Matter?” (available here) puts some numbers and analysis behind intuition.

Becker and Hvide look at roughly 65,000 Norwegian firms established between 1999 and 2007, in particular studying firms where the entrepreneur dies before 2009 (so firms are between zero and eleven years old in this sample). In this case, an entrepreneur is defined as an owner with at least 50 percent ownership, though most of the analysis focuses on entrepreneurs with more than 50 percent ownership (341 instances).

They find that firm performance drops significantly after an entrepreneur’s death relative to other similar firms that do not experience a death event: on average, four years after an entrepreneur’s death, sales drop by about 60 percent, employment drops by about 17 percent, and firms have 20 percent lower survival rates two years after the entrepreneur’s death. The effects appear to be long lasting and firms show no signs of recovering from them.

We can say with some certainty that there is causality here—that the entrepreneurs have a causal effect on the growth of the firm—because the authors conduct a series of robustness checks to address many potential questions about the strength of their findings.

No significant differences:

• Are the negative effects of an entrepreneur’s death only for firms that were performing poorly and would have had poor performance anyways? No, the effects occur for firms across all levels of pre-death performance.

• Are the negative effects of an entrepreneur’s death temporary (i.e. a period of turbulence and then return to normal)? No, the immediate effects are small relative to sustained effects that accrue over time.

• Do more mature firms fare better; does the age of firm matter? No, the negative effects are largely independent of the age of the firm, although more mature firms fare slightly better with survival rates.

• Does the age or gender of the founder matter; does the death of an older founder (60 years or older at founding) have less of effect? No significant differences are found.

• Do family firms fare better or worse; does it matter if the entrepreneur was married? No significant differences are found.

• Do firms in urban areas, with a greater and denser supply of entrepreneurs, which could serve as replacements, differ from firms in rural areas? No.

• Does the size of the firm matter? Only minor differences are found for small firms relative to large firms.

• Is the reverse happening: does poor firm performance lead to the entrepreneur’s death? The authors do not have data about the health of the entrepreneur or cause of death. However, they can compare the pre-death performance between firms that experience a death and those that do not and find there are no pre-death differences in performance. This suggests that death comes unexpected or that health issues are not significant enough to affect firm performance.

Some differences:

• Does the firm having team ownership matter? Yes, negative effects of the death of an entrepreneur still occur but are lessened. When Becker and Hvide lower look at owners with just a 50 percent share (204 instances) or a less than 50 percent share (495 instances), they find similar negative effects, but just less severe. For 50 percent owners, the effects are about half as large relative to greater than 50 percent owners (majority owners; the main analysis); for less than 50 percent minority owners, the effects are quarter the size of the case of majority owners.

• The loss of greater levels of human capital has a more pronounced negative effect. The death of entrepreneurs that are highly educated (more than 12 years of formal education) have greater negative effects relative to other entrepreneurs; firms that are in sectors that have on average a higher education level experience greater negative effects from a death event relative to firms in other sectors.

Getting back to the main findings, the authors do not know why there is such a stark difference between the drops in sales (60 percent) versus employment (only 17 percent). They suggest as one possibility that the entrepreneur has spillover effects on employee productivity which would hurt sales proportionately more (i.e. leadership in sales, or was a great salesperson), but don’t have strong evidence to present. This will be an area for future research.

I think this research serves as a precautionary tale to startups that will unfortunately face unexpected life events. Besides an entrepreneur’s death, other unexpected life events could see an entrepreneur suddenly removed from their firms. I am not convinced this research extrapolates to founder removal due to internal firm conflicts where a decision is made (either by majority of co-founders or investors), because those decisions are made by choice.

Give Me Your Entrepreneurs, Your Innovators

Dane and I have a policy brief out today that looks at a few possible scenarios of job creation that could result from a startup visa (paper is here; press release summary is here). The Startup Act 3.0 bill proposes to create 75,000 visas for immigrant entrepreneurs who found companies. We estimate this could result in, at a minimum, somewhere between 500,000 and 1.6 million jobs created after 10 years. The legislation calls for 75,000 visas to be active at any given time, meaning that 75,000 visas could be filled and then no new visas would be issued until one of those slots opened (presumably because a startup failed or the startup graduates from the visa program). Employment projections would be much larger if the pool was expanded or if visas were allocated annually. More details and caveats about how we constructed the scenarios are available in the paper.