3 ms·

Multiple studies have shown that startups are dying in the USA. The great era of startups in the USA was in the mid 20th Century. At least since the 1970s, new

Multiple studies have shown that startups are dying in the USA. The great era of startups in the USA was in the mid 20th Century. At least since the 1970s, new business formation has been dying: http://econweb.umd.edu/~haltiwan/dhjm_jep_5_17_2013.pdf http://econweb.umd.edu/~haltiwan/dhjm_jep_5_17_2013.pdf http://www.brookings.edu/~/media/research/files/papers/2014/05/declining%20business%20dynamism%20litan/declining_business_dynamism_hathaway_litan.pdf http://www.brookings.edu/~/media/research/files/papers/2014/... Here is the abstract of that last study: "Business dynamism is the process by which firms continually are born, fail, expand, and contract, as some jobs are created, others are destroyed, and others still are turned over. Research has firmly established that this dynamic process is vital to productivity and sustained economic growth. Entrepreneurs play a critical role in this process, and in net job creation. But recent research shows that dynamism is slowing down. Business churning and new firm formations have been on a persistent decline during the last few decades, and the pace of net job creation has been subdued. This decline has been documented across a broad range of sectors in the U.S. economy, even in high-tech." And here is the start of the first: "The pace of business dynamism in the U.S. has declined over recent decades. The decline in business dynamism is evident in a pronounced declining trend in the pace of both gross job creation and gross job destruction. An important component of these declining trends has been the decline in the firm startup rate. The decline in the startup rate has yielded a significant decline in the share of employment accounted for by young firms – this share has declined by almost 30 percent over the last 30 years. " And please, please, please note that a country can have an incredibly dynamic, innovative industry, but that nation can still be in long-term decline. See here for details: http://www.smashcompany.com/business/if-the-usa-is-the-most-innovative-nation-on-earth-then-how-can-it-have-a-period-of-economic-decline http://www.smashcompany.com/business/if-the-usa-is-the-most-...

jessriedel 11y ago
Are those studies equating "starting a small business" with "startup"? Paul Graham only uses "startup" to refer to business that start small but grow large is a short time, as he says in the OP article itself (and elsewhere). If you equate those two ideas, you won't be able to talk clear about this topic.
- lkrubner 11y ago
  When he refers to "the spread of computing power" he seems to be talking about software startups in particular. But when he says "The ultimate way to get market price is to work for yourself, by starting your own company" he is either talking about every imaginable form of business, or he is talking complete gibberish. If everyone goes into one sector, the economy would soon suffer an abundance in that sector, and a relative lack in other sectors. So his point is either true of any business that one might start, or he has no point at all.
  - ericd 11y ago
    There are "tech startups" that serve lots of "non-tech" sectors of the economy. One of the major beliefs in SV is that this is going to happen more and more, with companies making tech that augments or replaces entire sectors of human work. There's obviously a ways to go before this will be possible with some areas.
  - nostrademons 11y ago
    It's not unreasonable for everyone to go into one sector, and then for that sector to eat the entire rest of the economy (a la a16z's "software is eating the world" investment thesis). Imagine that software is an innovation akin to assembly lines. If you started a company in 1840 that used assembly lines, you were going into a very specific industry: textiles. If you started a company in 1890 that used assembly lines, it could be one of dozens of industries: textiles, steel, oil, tobacco, meatpacking, etc. If you started a company in 1930 that didn't use assembly lines, you were insane and destined for failure. What's happened, historically, is that as a technology diffuses into new ways of doing things, old industries simply die out, and then the new industry differentiates into replacements for it. Specialized production in the middle ages was divided into guilds (many of which are family names now): you had smiths, cobblers, bakers, butchers, weavers, etc. When the industrial revolution happened, there was simply "industry"; it was only later that this differentiated into the steel industry to replace smiths, the meatpacking industry to replace butchers, the textile industry to replace weavers, etc. Similarly, when software got started in the 50s, there was only the "software industry". Now, 60 years later, it's differentiating into search, social, messaging, e-commerce, developer tools, homesharing, adtech, fintech, ridesharing, delivery, etc. And it's likely there's more to come: if I had to locate us in the history of the industrial revolution, I'd say we're around 1890 or 1900. At that point, the great industries of the mechanical age (mass-produced automobiles, airlines, plastics, radio, television, recorded music) had yet to be invented.

jessriedel 11y ago

Are those studies equating "starting a small business" with "startup"? Paul Graham only uses "startup" to refer to business that start small but grow large is a short time, as he says in the OP article itself (and elsewhere). If you equate those two ideas, you won't be able to talk clear about this topic.

lkrubner 11y ago
When he refers to "the spread of computing power" he seems to be talking about software startups in particular. But when he says "The ultimate way to get market price is to work for yourself, by starting your own company" he is either talking about every imaginable form of business, or he is talking complete gibberish. If everyone goes into one sector, the economy would soon suffer an abundance in that sector, and a relative lack in other sectors. So his point is either true of any business that one might start, or he has no point at all.
- ericd 11y ago
  There are "tech startups" that serve lots of "non-tech" sectors of the economy. One of the major beliefs in SV is that this is going to happen more and more, with companies making tech that augments or replaces entire sectors of human work. There's obviously a ways to go before this will be possible with some areas.
- nostrademons 11y ago
  It's not unreasonable for everyone to go into one sector, and then for that sector to eat the entire rest of the economy (a la a16z's "software is eating the world" investment thesis). Imagine that software is an innovation akin to assembly lines. If you started a company in 1840 that used assembly lines, you were going into a very specific industry: textiles. If you started a company in 1890 that used assembly lines, it could be one of dozens of industries: textiles, steel, oil, tobacco, meatpacking, etc. If you started a company in 1930 that didn't use assembly lines, you were insane and destined for failure. What's happened, historically, is that as a technology diffuses into new ways of doing things, old industries simply die out, and then the new industry differentiates into replacements for it. Specialized production in the middle ages was divided into guilds (many of which are family names now): you had smiths, cobblers, bakers, butchers, weavers, etc. When the industrial revolution happened, there was simply "industry"; it was only later that this differentiated into the steel industry to replace smiths, the meatpacking industry to replace butchers, the textile industry to replace weavers, etc. Similarly, when software got started in the 50s, there was only the "software industry". Now, 60 years later, it's differentiating into search, social, messaging, e-commerce, developer tools, homesharing, adtech, fintech, ridesharing, delivery, etc. And it's likely there's more to come: if I had to locate us in the history of the industrial revolution, I'd say we're around 1890 or 1900. At that point, the great industries of the mechanical age (mass-produced automobiles, airlines, plastics, radio, television, recorded music) had yet to be invented.