10/11/2010 8:00:00 AM By
Suggestions from Brookings and others were successful in getting the Small Business Administration to add a section to its Strategic Plan
recognizing the importance of data:
Strategic Objective 3.3: Promote the availability, analysis, and dissemination of the most current, accurate, and detailed statistics possible on small business.
1. Advocate for improved data collection on small business activity. Pursue new avenues for improved and expanded data products on small business by working with other government agencies and external sources.
2. Carryout and publish data research and analysis. Through both internal analysis and contract research, publish regular, useful, high-quality data and indicators on small businesses and the role that they play in the economy.
3. Raise awareness of data and findings. Publicize the availability of data and findings to federal agencies, Congress, small business organizations, research organizations, the media, and other stakeholders.
Original Post - September 10, 2010
As an organization advocating for more information, the Small Business Administration (SBA) is lacking. They are not a U.S. statistical agency, meaning they don’t actually collect any of their own data. SBA is left to seek different cuts of Census, BLS, or other data by business size or to hope they collect relevant new data on small business. While this has been effective for the SBA to a point, their strategies-to-date seem ineffective at driving consistent, long-term collection of surveys or data that are on topics unique to small business. The Survey on Small Business Finance, canceled some years ago by the Federal Reserve
, would have been much more conceptually at home in the SBA or in an existing statistical agency with strong SBA support. Instead, it, like other topics unique to small business have only been implemented in a hodgepodge manner.
Why do I bring this up? Well, recently the Small Business Administration put out for comment its “Strategic Plan: Fiscal Years 2011- 2016.” I was woefully negligent in actually getting my comments submitted formally, but I wanted to offer a quick nod to comments submitted by Andrew Reamer, formerly of the Brookings Institution and now at George Washington. Andy very aptly points out that data and statistics are missing from the vision for the next five years. If these topics remain off of SBA’s formal radar then any advances in data collection on small businesses that are nascent within the minds of statisticians at Census or BLS will likely only remain thoughts. Adding such concrete recommendations to a document like this can set up future funding efforts or intrapreneurs at one of the agencies. I know from my work that there are a lot of people in the statistical offices looking for new and innovative products. I hope the SBA can help to be an outside advocate for them.
10/7/2010 9:00:00 AM By
A couple of weeks ago I was in Paris for a great meeting at HEC on University Entrepreneurship organized by Thomas Astebro
. That left me with time for a morning status meeting with Mariarosa Lunati who is currently heading up the Organisation for Economic Co-operation and Development's (OECD) Entrepreneurship Indicators Project (EIP). EIP is a program that Kauffman helped to launch more than five years ago
but which has only in recent years started to produce statistics
that are harmonized across countries on important topics such as the number of high-growth companies, gazelles, and business births/deaths.
The OECD continues to push to get more countries into the series. Through their partnership with Eurostat, the data collection and production of the harmonized statistics has become mandatory within the European Union meaning that within two years all members of the European Union that have the technical capability to produce the EIP statistics will do so annually. While the U.K. might produce statistics this year, that will bring countries such as Germany into the fold. Many non-OECD countries are also now producing the statistics – namely Brazil – with Australia hopefully coming on line this year and Japan in 2012.
In November, EIP will produce two statistical briefs. One of the briefs is sure to be on timely indicators of entrepreneurship across countries while the other topic is still in discussion. These two pieces will be distributed widely and bring some focus to the topic of entrepreneurship before and during Global Entrepreneurship Week. At that same time, the OECD expects to post on its website updated Excel files of country statistics gathered. A longer, more in-depth report will be released in the spring of 2011. In this report, the OECD also expects to discuss hot topics in entrepreneurship measurement including green entrepreneurship, women’s entrepreneurship, and migrant entrepreneurship.
By all indications the EIP is attempting to develop a core set of indicators which are on their way to becoming good time series, as well as to remain relevant in discussing topical issues that can contribute to other OECD programs or identified needs. In this regard, I tip my hat to Mariarosa for the great work she is doing. It’s very exciting to see so many positive developments from this program. With the increasing number of years and countries available, I think the next few years should see much wider analysis of the data within the academic community.
Read the OECD's full summary report
of recent activities.
10/7/2010 5:47:55 AM By
9/9/2010 9:00:00 AM By
At the 2010 Academy of Management workshop, we helped to organized a packed workshop for people looking to use more longitudinal data in their research. The following are the PPTs that were presented at that workshop. In particular, I really enjoyed Jon Eckardt's discussion of using internet data for entrepreneurship research.
9/8/2010 9:00:00 AM By
Last fall I participated in a really unique workshop at Yale put on by one of our Kauffman legal fellows, Victoria Stodden
. The purpose of the workshop was to discuss data and code sharing best practices and issues for creating replicable research. While the workshop was a bit more in the computational science space than I am fully comfortable, I found the conversation incredible and the goals of the effort beyond compelling. What has resulted is a Data and Code Sharing Declaration (just published in IEEE Computing in Science and Engineering
). This is a document that should be taken up for discussion at foundation and other funder events, in policy circles, and within the scientific academia as it lays out early and clear recommendations for actions each group can take to further data sharing and replication in the future. It is a document which anyone who curates data, journal editors, and all scientists should be discussing.
9/7/2010 3:00:00 PM By
I am really pleased to announce that the Kauffman Foundation has awarded a grant to Ohio State University to extend data collected in the 1979 National Longitudinal Survey of Youth (NLSY79)
. The NLSY79 is among the most widely used surveys in the social sciences, with a constituency that spans public health, social work, psychology, sociology, demography, economics, geography, financial planning, and public policy. The NLSY79 began in 1979 with a nationally representative sample of 12,686 individuals born in 1957-64. These individuals were surveyed annually from 1979 to 1994 and biennially from 1994 onward. Almost 8,000 respondents—80% of the eligible sample, given that some respondents died or were intentionally dropped—participated in the latest round. Most respondents have been followed from full-time schooling through the first decades of their work lives, and have reported detailed information on virtually every employer encountered along the way. As a result, the NLSY79 is a preeminent data source for researchers and policy makers seeking to understand school-to-work transitions, patterns of employment and nonemployment, job and occupational mobility, wage growth, gender and race differences in labor market outcomes, and much more. Self-employment is among the topics for which the NLSY79 has been widely used, given that respondents identify the “class” of worker (government; private, for-profit company; non-profit organization; self-employed; or working in family business) for every job reported.
Longitudinal surveys are not changed without a great deal of debate and consideration and yet the NLSY79 has reacted to the groundswell of desire among researchers for more data on entrepreneurship to add an entire module on business ownership to its 24th round of collection. Most importantly, NLSY79 is going beyond just asking about business ownership in the prior 2-year period, which is their typical practice for most topics, and gathering data on all business ownership experiences over the previous several decades. What I have described to this point the staff overseeing the NLSY79 has been able to squeeze out of their existing contract to administer the survey on behalf of Bureau of Labor Statistics.
Kauffman funding will support two additional activities designed to increase the usefulness of the data:
Task 1: the full linking of the retrospectively-collected data across the many rounds of collection to all relevant previously-collected data
Task 2: a follow-up question in 2012 to clarify any discrepancies in the responses of respondents over time.
Retrospective collection is very difficult. Based on cases that have already been completed in 2010, OSU projects that (i) about 1,500 respondents (20% of the sample) will report business ownership in the 2010 module; (ii) roughly 1,100 non-current businesses will be reported by these respondents; and (iii) as many as 700 of these non-current businesses will merit follow-up questions to confirm linkages with previously-reported jobs. Correcting the data administratively and proactively will jump start research in this area and eliminate gray areas of responses that otherwise would be thrown out.
No other major longitudinal population surveys are underway in the U.S. which could be considered real alternatives to this type of collection. While I have for many years tried to get some updates going to the PSID
design, I am not aware of any current or prospective efforts at PSID to expand their collection of entrepreneurship data.
While work is underway at OSU on this collection, the actual data from 2010 won’t be available until 2011. Additionally, the clarifications on responses will not be available until 2013. Thus, we are pleased to make this announcement and hope that it will start to get the research community excited about using these data when available. If you would like to be notified about the data’s release, please add a comment to this blog posting. For now that is our best way of keeping track of interested potential researchers.
8/30/2010 9:00:00 AM By
Apologies to my readers as I’ve been an absent blogger for the last couple of weeks with other work priorities. But it’s exciting that I’ve only been absent a couple of weeks and there is so much to update on regarding entrepreneurship and innovation data! So, I hope to catch up on postings in the coming week.
One of the biggest data releases of the year is a new series from the Bureau of Labor Statistics
. This is an extension of their population Business Employment Dynamics but one that adds more detail on two important measures – business age and survival – into their publically available series. Dive into the data
or read their overview of the release
I plan to spend more time with this data in the next few weeks as it fits nicely into a paper I am writing. For the time being, I looked at the data mostly to see if there was anything that looked really off or significantly different from a similar data release from the Census Bureau
and I am pleased to not have many problems to report. Perhaps the table that I found of most interest employment in surviving establishments. BLS has done an outstanding job of presenting the data in a logical manner so that each cohort of establishments can be followed over time without difficulty. So with this perspective
, it becomes easy to see that on average after more than 15 years, only about a quarter of establishments are surviving but that these establishments have grown in employment over time from an average of 7 to about 17 employees.
The BLS release of this data continues a trend at U.S. statistical agencies to add age as a component of ongoing business measurement and public release. It's a very exciting trend to see continuing and should bring more research in this area. Also of note in the last few weeks on this topic is the publication of a "Who Creates Jobs? Small v.s Large vs. Young"
by Haltiwanger, Jarmin, and Miranda. This is a good overview of why looking at business age as a component of analysis in government data is an important issue when studying job creation.
I should note that this is establishment data – not firm level data – so it doesn’t match perfectly with Census in that regard and will include both new companies and new locations of existing companies.
7/22/2010 9:52:15 AM By
Where do you place your confidence? Government, private business, newspapers? Gallup has an annual Confidence in Institutions Index which has been tracking a very similar question back to the 1970s. Today they released an update for July 2010
. While the headline numbers are mostly about confidence in Congress (ouch - 11 percent) and the Presidency (down 15 percent in a year), what always strikes me in this research is how difference small business and big business are rated. In 2010, when asked "Now I am going to read you a list of institutions in American society. Please tell me how much confience you, yourself, have in each one -- a great deal, quite a lot, some, or very little?," the second highest response in terms of confidence was small business with 66 percent while big business was scraping near the bottom with 19 percent confidence.
In a survey of youth in 2007 (and actually in earlier iterations), we ask a somewhat similar question, "How much do the following help make your community a better place?" More details on the survey
available online but I don't think we published these tables broadly.
||Not at all
We are in the field with an update to this youth survey so we'll see how some of these things have changed among America's youth in the last three years. In comparing the two questions and populations, it's interesting to see the only group the youth tended to have different views of in comparison to ther others is the role that individual people play in making their community a better place. While I realize the phrasing of the two questions is different - confidence vs. making community a better place - I would argue that the two questions really are trying to get at similar things. While Gallup doesn't include individuals in their categories (as they really aren't institutions), it looks to me on reading that some of the confidence that adults might have in small business is really a confidence in the power of individuals. Additionally, it looks like the conception of small business being different from big business is something which must emerge after youth.
I had a great conversation with Carol Corrado at the Conference Board last week in the neighborhood of this topic that was sparked by a Business Week article
and some responses from colleagues/grantees at Kauffman
. I'm not going to jump into that full discussion but the thing I did want to raise with my post here is that there is no doubt that both small and big business play important roles in the U.S. economy, indeed wealth creation can only come through private businesses. Carol's point, I think, was that we really don't have a fully-conceptualized picture of the roles, particular how big business is often supportive of small business. I won't dispute that (and I hope our data infrastructures and conceptualizations can continue to advance so that these questions can be quantified). But the polling continues to show the picture people get when thinking about the little guy is very different than when thinking about big business.
7/20/2010 8:00:00 AM By
As a follow-up to my post yesterday on current trends in entrepreneurship in the U.S., I wanted to highlight another new report available - the Survey of Business Owners (SBO) 2007
. While this is not a particularly timely report, it is a deep report, bringing us our best snapshot of U.S. businesses by demographic once every five years. While the SBO has a varied release schedule for reports with many more forthcoming, what came out recently was new data on ownership by gender and race. My reaction to the report would fall very close to my colleague Alicia Robb's, which was summarized in a recent Business Week article
- it's great to see some growth in the numbers of businesses but extremely troubling to see where the growth is occurring and the many areas of continued lag.
This graphic from Bloomberg captures two things in particular 1) we are seeing growth in numbers of minority businesses, almost all in non-employer businesses and 2) the growth in minority-business ownership is dwarfing growth in non-minority firms. BUT staggering levels of discrepancy in terms of lacking larger minority-owned businesses remains mostly intact. In other work from the Kauffman Index
, we have shown that the nature of those entering entrepreneurship has changed greatly in the last decade, partly by propensity and partly by changing population demographics. The SBO report doesn't get into the changing population demographic, which I suspect is driving the change in the Hispanic numbers in the report but can't be seen as a driver of the large increase in Black-owned businesses.
As for trends with women-owned businesses, in 2007, there were 911,285 women-owned employer firms with 7.6 million persons employed and total payroll of $218.1 billion. This is an increase of 6.2 percent and 25.7 percent respectively from 2002. This compares to 3.2 million men-owned employer firms that employed 41.6 million persons with a total payroll of $1.5 trillion. I hate to be a pessimist but given how integrated, educated, and a part of all economic life that women are today, these are numbers that have to change faster. There is no reason we should still be seeing women's employer businesses averaging 8 employees while male-owned firms are averaging 13 employees.
The Census Bureau now has within its grasp the data infrastructure to produce annual data updates on these statistics of business owners by demographic. It absolutely should do so. Without more frequent updates on these trends I fear efforts to create more minority and women's employer businesses will never gain traction. There remains a great deal of untapped economic potential in the disparate levels of minority and women's businesses.
7/19/2010 3:00:00 PM By
The Challenger Job Market Index
, a quarterly survey of approximately 3,000 job seekers, was out today with some new info on 2010 transitions into entrepreneurship. It is surprising in that it reports a sizable decline in entry into business ownership in the first half of 2010, with only 3.7 percent of their sample opting to start their own business (down from 7.6 percent in the first half of 2009 and 9.6 over the last two quarters of 2009).
The natural question here is what does this all mean when put up to other statistics currently available? Certainly a decline like this is not a positive development since it shows about the lowest rate of start-up activity in this survey since Challenger began collecting it (see related post by a colleague showing historic trends
). But at a time when job seekers are seeing record lengths of long-term unemployment
and are likely very financially constrained, I can’t say that I am entirely surprised to see a decline in the percentage of the unemployed who are seeking other forms of employment besides starting their own business. I think the press release from Challenger
rightly identifies some of the opportunities this group might be facing as hiring has eased with existing employers and things like the credit crunch in lending.
But do we expect to see a drop of half
in the rate of new business formations in 2010? No. The latest data we have from the Kauffman Index
which goes through December 2009 shows increasing rates of households reporting entry into entrepreneurship and as we’ve been documenting in other recent reports
, many rates of entrepreneurship seem to be extremely steady. What I wouldn’t be surprised to see in new data coming out on 2009 and 2010 is a moderate decline in employer business formations. I expect these declines looking at trends from the Bureau of Labor Statistics
or other not yet published leading indicators collected by the World Bank. But a 50 percent decline has never been seen in government data collected on this topic in the United States.
There is little doubt that when taken on the whole that there is an increase in necessity entrepreneurship and a possible
decline in opportunity-driven entrepreneurship in the U.S. currently. We’ll continue to look into these things more in the coming months in new papers in our Kauffman series on this topic, but the real question which we can’t answer currently is if there are new transformational companies being born today, as my colleague Dane Stangler has shown has occurred in past downturns
I’ll just end with a call for all my federal statistical agency colleagues to redouble the effort to get some more timely indicators on new business formation. BLS currently leads the way with their BED
and CPS-based statistics
but I know Census is considering some new monthly measures and I suspect something similar might be possible at BLS.