05 March 2017

The Fourth Economy & the Popularization of Entrepreneurship (or how work evolves from farming to entrepreneurship)

Graphic created by Jacob Morch jacobmorch.com

The definition of work changes as economies evolve. The grandchildren of farmers became factory workers and the grandchildren of factory workers became knowledge workers.  There’s good reason to believe that the definition of employee will change again, this time into something like entrepreneurship.

Thomas Jefferson imagined the United States as a country of educated, gentleman farmers. Even when he became president in 1801, though, the percentage of Americans farming had begun its steady decline. Now, each month economists await the announcement of nonfarm payroll employment. Today farm jobs are not even included in the country’s defining measure of jobs lost and gained.

Alexander Hamilton’s vision of an industrialized nation turned out to be more prescient but in recent decades, manufacturing’s share of the work force has also been in steady decline. Next century, economists may await the nonmanufacturing payroll employment report.

Millions voted for Trump and his promise to bring back manufacturing jobs. As promises go, it seems more akin to a 1916 campaign promise to bring back farming jobs than an adaptation to new realities. Yet acknowledging that farming and manufacturing are unlikely to reverse their decline leaves us with the question about the source of next generation jobs.
The economy has shifted but policy has not. Until economic policy begins to address the new limit, it will continue to be ineffective.

Over the last 40 or 50 years the per capita GDP growth rate has fallen. The fallout is not just economic. It has made voters less trusting of major institutions and expressed itself in surprising victories for BREXIT and Trump. Most people now feel that “the system is broken, unfair, and failing them.”[1]

Meanwhile, one place that has done remarkably well in the last half century is Silicon Valley, a place that more than any other has become synonymous with entrepreneurship.

About a century ago, Henry Ford made business history by doubling the wages of his factory workers. Doubling. Not only was he making cars more affordable, he was paying working class people enough to buy them.

In 2016, median wages in the US were about $51,000 a year. Like Ford, Silicon Valley has doubled that. In Santa Clara County – one reasonable approximation of Silicon Valley – average wages were $117k, or 118% higher than the national average.

It’s possible that Silicon Valley is an anomaly, a place that other communities can only envy but never emulate. A more interesting possibility is that Silicon Valley is to a new entrepreneurial economy what Manchester, England of the 1700s was to a new industrial economy: just the first place to enter a new economy whose practices will eventually spread around the world.

Four Economies and Four Limits
Agricultural economies give way to industrial economies, which give way to information economies. Most people share that intuition but their understanding of what these labels mean and how to distinguish between them is fuzzy. Even industrial economies have farms and information economies have factories. It takes a little explanation, but limits can clarify the distinction between different economies and predict a fourth, entrepreneurial economy.

Period in West
1st, Agricultural
1300 to 1700
2nd, Industrial
1700 to 1900
3rd, Information
1900 to 2000
4th, Entrepreneurial
2000 to ~

Before talking about economies, imagine a factory with four stages. It gets raw materials in on one end and sends product out the other. The materials that become a finished product must pass through all four stages before they’re sold.

The numbers and height of the bar indicate how many products a stage can process in an hour. The first stage can process only 1, the second can process 2 and the fourth and final stage has the capacity to process 4 products an hour.
The customers don’t buy the unfinished product from any intermediate phase, though. They only buy product that comes out of the whole factory, product that has passed through all four phases. The question is, what is the capacity of this whole factory? How many products can it produce per hour?

The answer is 1 per hour. Your factory’s capacity is equal to the capacity of your first stage. You could call that a bottleneck, a constraint or limit. Whatever you call it, this limit defines the capacity for your whole factory. If it can only feed the next stage 1 item per hour, it doesn’t matter that the second stage has the capacity to process 2 items per hour because it won’t get product fast enough to process that many.

Until you increase the capacity of the first stage, you will not increase the capacity of your factory. So, you experiment. Maybe you speed up the process, simplify the process or just buy a second machine for that first stage. However you do it, you eventually double the capacity of this first stage to get a picture like this:

The good news is that by doubling the capacity of the first stage you have just doubled output for the whole factory. Armed with the knowledge that focusing on the first stage makes all the difference, you continue to experiment and invest in improving that first stage until you find a way to double its capacity again.

This time, though, doubling the capacity of your first stage does not change your factory output. Why? You were so successful at improving the first stage that it is no longer the limit to your factory. Your limit has shifted elsewhere.

Two lessons from your factory could apply to any system.[2]
  •        To improve the system, you have to focus on the limit, and
  •        Success eventually shifts the limit.

So, what limits an economy? In every introductory economics course, students learn that there are just four factors of production: land, capital, labor, and entrepreneurship. Anything of value created by an economy depends on some mix of these four factors and one of those would have to be the limit at any given stage of economic development. Land includes all natural resources, from herring to oil, acreage and cotton. Capital includes the financial and industrial tools that transform those natural resources into finished products, the factories that can turn cotton into clothing and the stocks or bonds that finance the machines and factories. After the industrial revolution, the labor of knowledge workers – people like accountants, engineers and advertisers – who manipulate the symbols of things rather than actual things was the most defining labor. Finally, entrepreneurship brings together land, capital and labor into a profitable enterprise.

The four phases of a factory can become four factors of production in an economy and we can examine limits to an economy in the same way that we examined limits to the factory. The output of an economy can be measured by things like jobs or wealth, income or GDP.

Different limits create different economies
Agricultural economies are limited by land. Wealth between 1300 and 1700 didn’t result from advances in information technology (not that the Gutenberg Press wasn’t disruptive) but instead came from trade, conquest, and colonization with faraway lands and creating nation-states and private property in your own land.

An industrial economy is limited by capital. Between 1700 and 1900, the creation of wealth was less about exploration, conquest and colonization than it was about building the factories that could turn raw materials into finished goods and then build out canals and railroads to distribute those goods. Wool and cotton became fashion. Iron ore became railroads. Skyscrapers rose in cities and cars emerged to drive between them.

An information economy is limited by knowledge workers. Between 1900 and 2000, it wasn’t enough to have factories that could make more products than anyone had ever seen before. They had to be the right products (which required marketing and design expertise) made for and sent to the right places (which took manufacturing and distribution knowledge) by the right methods (which took advertising and retail display experts.)

An information economy emerges after an industrial economy. Before the automation of the industrial economy, you need workers to manipulate actual things, afterwards, machines can do that and  labor can shift its focus to manipulating symbols. The sequence from agricultural to industrial to information economies is not just an historical sequence, it’s a logical one.

Period in West
1st, Agricultural
1300 to 1700
2nd, Industrial
1700 to 1900
3rd, Information
Knowledge Workers
1900 to 2000
4th, Entrepreneurial
2000 to ~

Economies are complicated and progress is slow so it makes sense that as communities gradually overcome limits they’ll cling to the processes that once made them great. Like the factory manager who keeps doubling the capacity of his first process step to no avail, communities can continue to create foreign colonies, spending huge sums on a global empire even after they’ve entered an industrial economy. Or more recently, they might pump money into their economy or create graduates past the point that capital or knowledge workers actually limit the rise in per capita GDP. It is almost inevitable that communities will continue to do what they’re now good at even after reaching a point of diminishing benefit. Cultures last longer than cost-benefit analysis and new practices become old habits.

An additional complication is that there are always pockets within a larger community that face earlier limits, and those limits define local culture and politics. When natural resources are the basis for wealth in a region, for instance, it will be more religious and more inclined towards policies like a strong military that support the notion of a zero-sum economy. It’s not the ingenuity of people that creates an oil field but is instead just a gift of God or nature. And that oil field doesn’t get larger because we decided to share it. Either I own it or you do, and rather than win-win we’re going to have a winner and a loser in this exchange. There will always be regions that lead or lag in development and thus will lead or lag in the reality they experience and that informs their convictions. It’s not just that a person living in rural Kentucky has a different political philosophy than her peer in Cambridge, MA; the daily reality that informs her perspective is different.
One other way to understand a limit is to look at its price. Scarce factors are expensive and abundant factors are cheap.

The success of the second economy made capital abundant. Traditional bankers who emerged from the second economy (many of our current banking practices were defined in England by 1900) carefully loaned out money, trying to minimize the risk of losing capital. Venture capitalists, by contrast, treat capital as abundant and fully expect to lose quite a few investments. Given they’re taking equity in a new firm rather than hoping to get back capital with interest, they know that only a fraction of their investments need to succeed in order for them to get great returns. Traditional banking evolved when capital was scarce: venture capitalists evolved when capital was abundant.
What is scarce now? Entrepreneurship and we can see that in its price. At 31, Bill Gates became the richest self-made billionaire in history. A generation later, Mark Zuckerberg became a billionaire at 24. The price of capital is the interest rate and towards the end of last year, investors owned about $12 trillion in negative interest rate bonds. Trillion. We have a glut of capital and a shortage of entrepreneurs, which suggests that effective policy would focus on increasing the supply of entrepreneurs rather than the supply of capital. Between 1700 and 1900, we learned how to increase the supply of capital through a variety of means, from popularizing savings and investment (from founding father proverbs like “A penny saved …” to expanding the number of people who bought wartime bonds and then later became savers) to changing the money supply or interest rates. If policy makers think that we’re short of capital, they can quickly pump billions into the economy. There are no comparable policy levers for increasing levels of entrepreneurship. Not yet.

When The Old Limits No Longer Limit
If capital were still a limit, we’d be in great shape. The S&P 500 have $1.5 trillion in cash and in the third quarter of last year they paid out $200 billion in dividends and stock buybacks. Banks excess reserves have dropped from their August 2014 high of $2.7 trillion but are still at a staggering $1.9 trillion.[3] (Before the Great Recession, excess reserves in the US were closer to $1.5 billion.)

Our education system helped us to overcome the limit of knowledge workers. In 1900, less than 10% of 14 to 17 year olds were formally enrolled in education. By 2000, less than 10% were not. In a century, the US went from an industrial economy dependent on child labor to an information economy dependent on adult education. That helped to transform life in the 20th century, real incomes increasing 6X to 8X and life expectancy rising from 47 to 77.

If knowledge workers and their information technology were still a limit, creating more graduates would help to create more jobs. In 2013, the American education system created 3.7 million graduates, everything from folks with AA degrees to PhDs and all the degrees in between. That same year, the economy ended the year with 2.4 million more jobs than it had at the start. We’re creating graduates faster than we’re creating jobs, 15 new graduates for every 10 net new jobs. It’s no wonder that student debt is becoming a growing issue.

It’s not just ineffectual to pursue old policies in a new economy. It can be dangerous.

A glut of money creates problems. Investors in search of returns, unwilling to accept negative interest rate bonds, too readily bought expensive things like tech stocks in 1999 or subprime mortgage instruments in 2007. Trillions in investments can create a series of bubbles and busts as it wanders the earth like a murmuration of starlings in search of returns.

A glut of graduates creates problems. Young people not only start careers with mounting debt but find it more difficult to find jobs they could not have worked with just a high school diploma. Millennials who are the best-educated generation in history nevertheless fear that they’ll be the first generation in American history to do worse than their parents. (This student debt will also make it tougher for them to finance startups. As medical school has become more expensive, for example, the percentage of doctors working for large groups or hospital has gone up relative to those who start a private practice.)

One consequence of continuing to pursue dated policies is that it makes it tougher to pursue any policy. When incomes are steadily rising, politics is civil. Families can pay a little more in taxes to support schools and help the poor while still taking home more pay after taxes. When incomes are stagnant, politics becomes more divisive. Few people like the idea of not supporting education or the sick but if the choice is between that or less take home pay? Well, the conversation becomes more heated and compromise is harder to reach on top of the fact that everyone starts this policy conversation disenchanted and bewildered.

We don’t need to jettison incredible financial and educational systems that are essentially over-producing, creating more capital or graduates than we can fully employ. We just have to stop looking to those systems as the means to create jobs and wealth. As we become successful at overcoming this new limit of entrepreneurship, we’ll be able to fully employ capital and college grads. Eventually, we will even create enough demand for them to bid their prices up further.

The Central Question of Every Economy
The central question for any generation concerned about economic progress is how to overcome its limit, not the limit of its grandparents or founding fathers. Creative answers to that question result in a new economy and a very different community.

In retrospect, the central question of economic development from about 1700 to 1900 was simple: how do we get more capital and make it more productive? The creative answers to this included everything from the Dutch stock market, Rothschild’s international bond market and the British banking system to the spinning jenny, steam engine, and continuous production technology. (The question is simple. The answers can be complicated.)

The central question of last century was, how do we create more knowledge workers and make them more productive? The creative answers to this included the popularization of K-12 education, the modern university, R&D labs, the modern corporation and information technology.

The question that policy makers everywhere – city hall and senate floor, corporate boardrooms and universities – should now ask has two parts:
  •       How do we create more entrepreneurs and make them more effective?
  •        How do we make employees more entrepreneurial?

Creative answers to these simple questions will transform the economy. We now have a financial system and an education system. We don’t have an entrepreneurial system but instead expect our entrepreneurs just to show up, like autodidacts in 1800. Changing will be an odd, fascinating and profitable project. Think about educating students to be prepared to become entrepreneurs in the same way that we now educate students to become university students and knowledge workers, for instance, or changing the definition of employee.

Changing the Definition of Work. Again.

Perhaps more interesting than the question of how to create more entrepreneurs is the question of how to make employees more entrepreneurial. We – rightfully – make a big deal about national economic policy. It’s worth keeping in mind that measured by GDP or revenue, of the 100 biggest economic entities only 31 are countries; the other 69 are corporations. (Walmart’s $480 billion in revenue would put it just between Sweden and Belgium’s GDP.) Corporate policy deserves as much discussion as national policy if we’re interested in progress. The most important topic in this discussion might be to ask what it means to be an employee in a time when AI like IBM’s Watson is liable to automate knowledge work in the same way that capital automated manual work.

Think about changing employment so that employees within a corporation had as much freedom to pursue new ventures as citizens within a country. Roughly 800,000 Americans make more than the $400,000 a year that we pay the president.[4] That sort of thing was unthinkable in Egypt under Hosni Mubarak or France under King Louis XIV, but as nation-states evolve, people within them have the potential to prosper more than even the head of state. Contrast that with how evolved the corporation is. While it’s common for professional athletes or portfolio managers to make more than their managers, it is rare that anyone inside a traditional Fortune 500 firm makes more than the CEO. What if employees could become more entrepreneurial, were able to create equity by taking existing products into new markets or by leading product and business development efforts that are akin to startup activities? And what if the success of those ventures could actually result in their making more than the head of the company in the same way that an American entrepreneur has the potential to make more than the American president? This dispersion of power and pay is just one way that the popularization of entrepreneurship will change the corporation.

Overcoming the limit of entrepreneurship will require and result in new legislation, new education, and new definitions of what it means to be an employee. As importantly, it will continue in a grand tradition of the west, doing for business what earlier economies did for religion, politics, and finance. That is, it will expand freedom for the individual. There is no way to make employees more entrepreneurial without giving them more freedom.

There are interesting examples of popularizing entrepreneurship within companies. Ricardo Semler did something interesting with his Brazilian company Semco. He gave his employees freedom to negotiate work arrangements. People working side by side on the factory floor doing similar work might have very different arrangements. One was paid hourly, another a monthly salary, another paid by piecework and another might actually be paying Semco to use equipment to make product that she – the employee – could later sell herself. Uber lets “employees” accept or reject specific fares and take just one fare a week or work all day. Amazon’s marketplace and Apple’s iTunes are platforms that let companies and entrepreneurs sell their own products. P&G is among the companies who richly reward successful product development leads whose responsibilities overlap quite a bit with entrepreneurs. All of these are examples of enabling entrepreneurship, blurring the boundary between traditional definitions of employee and entrepreneur, and giving the employee more freedom to define their own work and its results.

This matter of employees gaining more autonomy is not incidental to progress. Autonomy is a way to define progress and each new economy has given the individual in the West more freedom. If you have shoes you have more options about where to go than if you are barefoot; if you have a car you have even more options. If you live in a democracy, you have more options about what to believe and how to live than if you live in a theocracy. If you have a credit card you have more options than if you need to approach a banker to request a loan for a specific item, or can’t get a loan at all. If you have the freedom to create equity as an employee you have more freedom than if you’re expected to adhere to a process someone else defined.

The popularization of entrepreneurship will increase our product options and levels of wealth. Progress, though, is only partly about more and better products. That is only one way that our lives expand to include more options. The first economy didn’t just bring potatoes and tomatoes to Europe; it brought religious freedom. The second economy didn’t just bring fashion and automobiles to households; it brought democratic freedoms. And the third economy didn’t just give us radio and the polio vaccine; it made capitalists out of knowledge workers, giving them financial options that people in 1900 would have found as baffling as the internet. The fourth economy will transform business and work in the same way that the first three economies transformed religion, politics, and finance. That is, it will give us more autonomy, as economic progress always does.

As you might imagine, there is a great deal more to this new economy than would can be captured here. My book, The Fourth Economy: Inventing Western Civilization, can be found here. It's a longer read but it does explain progress from the Dark Ages to about 2050.

[1] https://twitter.com/Bill_Gross/status/821245915579240448
[2] Eli Goldratt, author of The Goal and Critical Chain popularized the ideas of Theory of Constraints (TOC) in the 1980s and 1990s within many Fortune 500 companies and government agencies.
[3] https://fred.stlouisfed.org/series/EXCSRESNS
[4] https://www.ssa.gov/cgi-bin/netcomp.cgi?year=2015

No comments: