I’m doing a marketing internship for a startup this summer and my boss sent me an article about the “lean startup”; this is an idea I heard about before but had no real interaction with until now. The lean startup is essentially a new methodology for starting businesses. It emphasizes testing hypotheses and responding to customer feedback with iteration and pivoting over the older methodology of forming a long-term business plan with “elaborative design” and “intuition,” as Steve Blank puts it in his article, “Why the Lean Start-up Changes Everything.”
I find the methodology very appealing. Attempting to make long term predictions about human choice among many different products appears absurd to me. Even with all the flaws in empirical testing to confirm hypotheses, the lean startup seems like a better method to use in the field of business (but not economics, as I explain in this article about the minimum wage).
However, I disagree with one point Blank makes in his article, a point made often elsewhere in articles about the lean startup. It is his belief that the introduction and spread of this startup methodology will have “profound economic consequences.”
A lower start-up failure rate could have profound economic consequences. Today the forces of disruption, globalization, and regulation are buffeting the economies of every country. Established industries are rapidly shedding jobs, many of which will never return. Employment growth in the 21st century will have to come from new ventures, so we all have vested interest in fostering an environment that helps them succeed, grow, and hire more workers. The creation of an innovation economy that’s driven by the rapid expansion of start-ups has never been more imperative.
Lean start-up techniques were initially designed to create fast-growing tech ventures. But I believe the concepts are equally valid for creating the Main Street small businesses that make up the bulk of the economy. If the entire universe of small business embraced them, I strongly suspect it would increase growth and efficiency, and have a direct and immediate impact on GDP and employment.
In this post, I’m going to focus on explaining my disagreement rather than trying to prove that it is correct. My view is based on Austrian economic theory, particularly Austrian business cycle theory, which I have set out in more detail elsewhere. I will explain the basics here (feel free to skip to part 5 if you already understand Austrian economics).
1) The concept of demonstrated preference:
Mises stated that individual humans act purposefully, choosing means to attain their chosen ends. The ability of humans to choose, means that humans have preferences. They prefer, and therefore choose, one end over another, and again prefer, and therefore choose, one set of means over another. When two individuals exchange with each other, each individual is showing that he prefers one good over the other. When Jim trades his house for Simon’s 300,000 dollars, Jim is demonstrating that he prefers $300,000 over his house, and Simon is demonstrating that he prefers the house over his $300,000. This is called demonstrated preference. Every time individuals act, they demonstrate a preference, just as displayed by the exchange between Jim and Simon.
2) Prices are predicated on demonstrated preferences:
As persons exchange, prices are formed. The price of the house in the previous exchange was 300,000. Likewise, we could say the price of a dollar was 1/300,000th of a house. As multiple men make exchanges among the same goods (houses and dollars), market prices are formed. Even most people who have never had an actual economics course have heard of supply and demand. But often they don’t understand the basis for supply and demand.
Supply and demand curves are just a graphical representation of a group of people’s real preferences. So maybe person A would be willing to buy 1 house at a price of 330,000, and 2 houses at a price of 300,0000. Maybe person B would be willing to buy 1 house only if the price was 300,000. And so on. These preferences added together would form the demand curve, stating how many houses (quantity on the x-axis) people together would demand at such and such prices (on the y-axis). The same thing would occur for the supply curve, being the preferences of sellers added together. In other words, it would state how many houses (quantity on the x-axis) people together would be willing to supply at such and such prices (on the y-axis).
3) The importance of the market clearing price:
Where the demand and supply curves meet and intersect is known as the market-clearing price. At this price, all buyers’ (“demanders”) and sellers’ (“suppliers”) preferences are satisfied. At any other price, there would either be more people wanting to buy than people wanting to sell, or vice versa. In other words, there would be a deficit of goods if the price was lower than the market clearing price or a surplus if the price was higher than the market clearing price. What brings it toward the market clearing price if exchanges occur at some other price? Arbitrage: the profitable reselling of goods (this isn’t that important for this post so if you don’t completely understand arbitrage, don’t worry about it).
4) Prices function as signals:
As we said before, prices change based on preferences, but there a number of other reasons that are derivative from preferences. For example, scarcity affects prices as well (through people’s preferences). If a good becomes more scarce, (people will value the remaining units more and) its price will go up. Considering this, prices are invaluable signals to individuals participating in the economy. For example, if a piece of machinery that a producer buys to make a product becomes more scarce and goes up in price, it is a signal to producers that they must use it more sparingly. Fewer can buy it now that its price is higher and marginal producers (those who barely profited from it before the price change) are driven out of the market as they now earn losses.
5) The government manipulates the interest rate, an invaluable price in a market economy:
When the government (and fractional reserve banking) prints money, it essentially increases the supply of loans (because they distribute this new money through the banking system by giving it out as loans), lowering the interest rate (for simplicity’s sake, the price of loans) below its free market level. This causes the business cycle. The interest rate like all prices is a signal. It represents the time preferences of individuals. If individuals want to consume more now and consume less in the future, they spend more and save less, and therefore lower the amount of money that they can loan out. By decreasing supply, this increases the interest rate. Likewise, if consumers want to consume less now and consume more in the future, they spend less and save more, and therefore increase the amount of money they can loan out. By increasing supply, this decreases the interest rate.
What happens when the government manipulates the interest rate? An exposition of complete Austrian business cycle theory goes beyond this post, but again, you can check that out here. Essentially though, now the manipulated interest rate functions as a bad signal. It does not represent the real preferences of individuals in the economy. If the interest rate is lower than what it would have been otherwise, it appears as if consumers are saving more (look at the last paragraph) when they really are not. Since the rate is cheaper, businesses can borrow more money to buy more capital goods (such as machinery; see the example in #4) to produce more goods for the future. But consumers don’t really want more goods in the future. They are spending on goods now and not saving enough money to buy those goods. As such, businesses invest in the wrong goods and start producing the wrong goods to sell at the wrong times.
Therefore, if the government does not allow price signals to work, it doesn’t matter what methodology businesses are using. They will not be able to improve the situation. If a signal is a bad signal, it doesn’t matter whether businesses are checking with it more often (as they do in the lean startup) or less often (as they do in the traditional startup methodology). In fact, if they are checking the bad signal more often, maybe they’ll make worse decisions and make the economy worse! (I say this somewhat jokingly, but certainly the reasoning is plausible)
What has to go are the structural impediments, the intervention in the economy by the government. That’s the only way the economy can recover. Then the lean startup methodology will be able to increase standards of living if it indeed is a better way of responding to consumer’s desires. But as long as the interest rate, the price representing the time preferences of individuals, is skewed, increasing the response rate to this incorrect signal will either have no effect, or at worst, be deleterious.
Two things I want to clarify about this post.
1) The word “fix” in the title might be the wrong word to use. The people I am responding to are not necessarily saying that this startup methodology will fix the economy, but they are saying it will have a noticeable effect on GDP and employment. The difference between that and fixing the economy is only one of degree. Regardless, I am disagreeing with that. I do not think it will have a significant impact in the current situation. If the structural impediments are removed, I do think it could have a significant impact on standards of living (and a common attempted measure of standards of living, GDP) and perhaps even employment, if producers are making less mistakes in judging what consumers desire.
2) A person might wonder why the interest rate is such an important signal. They might say, for example: Surely, if the price of potato chips is the price the government is manipulating, would it really have a huge impact on startups who are trying to sell different goods? Why is the interest rate any different? The reason the interest rate is so important is that it represents the time preferences of individuals. As such, any consumer decision made over time is relevant to producer decisions regarding the interest rate. If the interest rate is altered from what it should be, producer responses to supposed consumer preferences over time are now flawed. In addition, any signals correlated with the interest rate signal are flawed as well. So if a startup is looking at a slightly changed good that consumers want relative to the good they were producing, even if they’re not looking at the interest rate itself, this is still a flawed signal. As long as the consumer decision they are responding to is a decision with respect to time, the signal is flawed. There can be, of course, decisions made irrespective of time, such as if one product would never be desired, no matter what the period. But so many decisions depend on time or compare different goods with each other (often with a different period of production to make good 1 compared to good 2), more frequent producer responses to decisions irrespective of time would not have a very significant impact on GDP and employment.