Member preview

Gray Swan photo by Gareth Williams

Why the Economy Isn’t Growing Like It Used To

Secular stagnation is here to stay for the developed world, as we reach the limits to what one person can consume

The big economic trend to watch is secular stagnation. In a nutshell, the economies of the developed countries aren’t growing like they used to. Just one look at the 10-year moving average for GDP growth for the U.S. and other rich nations over 70 years, and it’s clear that something major is dampening growth. The cause is simple but counterintuitive: We are reaching the limit to how much a single human can consume. Without growth in consumption, there can be no economic growth.

One problem at the outset is the definition for secular stagnation. According to Wikipedia, stagnation is defined as follows:

a prolonged period of slow economic growth (traditionally measured in terms of the GDP growth), usually accompanied by high unemployment.

Economists added the word “secular” to indicate stagnation that can’t be explained by temporary causes, such as business cycles, but that aspect is already embedded in the word “prolonged”. A clearer phrase would be “long-term stagnation”, to indicate stagnation on the order of 20–40 years, longer than the Long Depression or the Great Depression. In other words, this is all about trying to figure out what happens when an economy never bounces back.

Former Treasury Secretary Larry Summers’s theory, which is responsible for the recent interest in secular stagnation, unfortunately returns to short-term thinking. He believes savings are too high, therefore the government needs to spur investment or inflation. But the government has been spurring investment in the form of low interest rates and quantitative easing, arguably since 1992 when the Federal funds rate was 3%. Secular stagnation is all about what happens when the treasury’s levers don’t work anymore.

Another popular theory is an old demographic one: Boomers are becoming a burden on society. The theory is appealing because it plays into our personal anxieties about the cost of elder care. And there is some logic too: The Boomers don’t work, so they must be a burden! But if anything, servicing the elderly should spur economic activity in the form of healthcare spending.

Related to the Boomer theory is the labor shortage theory, which is probably the most compelling one. Senator Marco Rubio said that we need more welders, and while his comment was more in the spirit of nostalgic political pandering, there is some truth to it. It’s too expensive to make stuff in the rich countries, so we offshore to China. In China it’s now getting too expensive to make stuff, so they’re starting to offshore to Indonesia or Thailand. Eventually it’ll be too expensive there, and they’ll offshore to Africa. Each time the offshoring happens, it leaves behind a country uninterested in craftsmen-like work.

Each time it happens, it reveals a growing leisure class within the middle-class, who is secure enough that they would rather gamble employable years in an attempt to find something more glamorous than being a nurse or an electrician. That security therefore implies a “good enough” level of consumption per household. Simply put, the rich countries aren’t hungry enough.


To get at the underlying cause of secular stagnation, we have to look beyond cyclical or demographic causes, and into the nature of economic growth as a whole, specifically Gross Domestic Product (GDP). Wall Street, the media, economists, and the Fed, all have agreed upon using GDP as the primary metric for economic activity. GDP is a giant, single number that has a clear dollar value, spelled out with this formula:

GDP = C + I + X – M + P
C = Consumption
I = Investments
X – M = Exports – Imports
P = Public Spending

The choice of terms in arbitrary, though, and the equation masks the interplay between all factors, specifically how much they are tied to the first variable: Consumption.

For example, the Investments variable is only important inasmuch as it leads to eventual consumption. Consumption is whatever goes into a human and disappears, whether it’s two hours of watching Game of Thrones or eating a piece of pie. But when a business buys a thousand Microsoft Office licenses, that counts as investment towards production. While it may look great when Microsoft’s stock subsequently goes up, it’s not the ultimate desired effect. Corporate America could theoretically swap random goods and services with each other ad infinitum, and investment stats would rise, but the consuming public would not benefit from this “economic activity”.

Exports – Imports is consumption that happens elsewhere, either in place or time. If we were to just control for trade, then any trade deficit spending for a country would have to occur from trade surpluses in the past. In which case, this is an exchange of consumption. One country consumed something from another country, and now the other country is returning the favor.

Public spending is also a distraction. If public spending is for goods and services that people wouldn’t buy on their own, then it’s fake economic activity, similar to the above corporate gift exchange. Or if public spending is for goods and services that people were going to buy anyways, then it’s just consumption by another party. After all this, we’re left with Consumption being the essence of GDP.

Moore’s Law is leveling off because of consumer demand, not because of physical limits

Emphasizing the importance of consumption in economic growth goes a little against the grain of traditional economic principles, which state that productivity gains are the ultimate, long-term driver of economic growth. A good example is Moore’s Law, which states that processor speeds double every two years. So if you’re in an industry that depends on computers, like complex manufacturing or assembly, and every two years your computing costs go down, so does the price of your goods, which means more goods that people can consume.

However, there’s a wrinkle in Moore’s Law. Processor speeds aren’t magically getting faster just because. Rather, gains in processor speed scale to research and development (R&D) budgets for chip manufacturers. R&D budgets can only grow if there is a growing market for those chips. So if the demand stopped, the R&D would stop, and processor speeds would stay the same.

Human demand is the tail that wags the dog of technology and productivity gains. We need stuff, which summons inventors to make said stuff. Capitalism means these makers compete with each other for efficiency. True, the more efficient they are, the more stuff we get. But it’s human demand that rules everything.

Sometimes an unexpected miracle in productivity gains comes along, like the Internet, but that has mainly improved the efficiency of the service sector, which obeys different rules, as opposed to manufacturing. Consider electricity’s impact on the Industrial Revolution versus the Internet’s impact on the Service Revolution. Electricity made the delivery of transformed raw materials — i.e. goods — radically more efficient. Any country that has undergone an Industrial Revolution has also undergone radical economic growth. While the Internet has sped up the delivery of services, though, subsequent economic growth has not materialized.

Economically, Japan’s Lost Generation is a preview for the rest of the developed world

Gains in productivity spring from gains in consumption, but consumption isn’t growing as fast anymore. Economists commonly look at household final consumption expenditure (HFCE) to measure consumption. Based on data available on The World Bank’s website, the U.S. has a 10-year average growth rate of around 1%, whereas in 1969 it was around 3%.

But these numbers are potentially over-inflated for our purposes because they include rent, which is up in developed countries. If instead of narrowly looking at GDP, we were to define good economic activity as a country making a lot of stuff and using a lot of stuff, then rent wouldn’t count. The land is still the same stuff, it’s just a whole lot more expensive. So if we account for the possible growth in rent, the consumption growth rate may actually be negative.

The same is true for other rich countries, some of which have gotten to that 1% consumption level earlier than others. Before the 1997 Asian financial crisis, Japan’s HFCE growth was around 3%. Since then it’s been less than 1%. Twenty years later, Japan’s GDP growth has averaged less than 1%. And Japan has had a mature, diversified economy, as well as some of the brightest economists and corporate managers, so they should have rebounded a few years after the crisis. Consumption and production should have more than prevailed over any down cycles or bubble bursts after twenty years. Japan’s Lost Generation is potentially a preview of what’s to come.

The developed world is reaching the limit of what it can consume due to the limits of human attention span

Part of the reason consumption has slowed down is because many productivity gains have only impacted services and soft goods. Soft goods, as opposed to hard goods, are a hybrid of goods and services. For example, a great episode of Game of Thrones requires a lot of services, as opposed to a Cadillac, which requires a lot of raw materials.

Hard goods can be made cheaper because of science and engineering. But how much more automated can the process of making a car get? How much cheaper can it ever be to produce a road? Productivity gains are supposed to lead to more consumption, which is economic growth, but we may be hovering close to the limit on productivity gains for hard goods.

When it comes to personal consumption, we have limited time and energy, plus we have limited needs. Going back to the car analogy, we could buy and discard a new car every day, but we don’t need to, nor do we want to. Theoretically we could stream ten TV shows at once, but again, we don’t have the time or energy to do so.

Perhaps our consumption could increase in complexity. An episode of Game of Thrones requires more economic activity to produce than a radio program, so in that sense, we have grown in our consumption of soft goods. But complexity isn’t the reason we buy things. Simple, cheap pleasures, like puzzle video games (Minesweeper, Bejeweled, et. al) are often all it takes to keep us occupied. So potentially, the complexity of what we consume, and the amount of engineering needed for basic and low-grade epicurean needs or wants, has plateaued, making the future similar to that depicted in Wall E.

The greatest growth in consumption is in the transition from the lower-class to the middle-class. Maybe someone in the lower-class rides the bus. As they move up, they buy a car. But as they continue to move up, they don’t buy more cars. We only “need” a used Honda Civic with A/C and power windows. By this metric, the middle-class has been content with their cars, TVs and white picket fences since the 1950s.

Even if the media spins the story of the shrinking middle-class, we see union membership down. We don’t see the middle-class taking to the streets with socialist agitation, like before the 1950s. They work for companies that have mountains of cash, and they aren’t asking, “Where is our share?”

But the middle-class is up to their eyes in debt, right? This is more of an issue of personal fiscal responsibility, which has stayed constant while the credit industry has grown in sophistication. Plus, student and housing debt have ballooned, which doesn’t count as consumption for our purposes. All that middle-class debt isn’t stopping people from buying all the TVs, cars, and iPhones they need — and nothing more.

The War on Drugs. The War on Poverty. The next “War” will be on Underconsumption

But if secular stagnation spreads to the rest of the world, which may happen 20–40 years from now, the implications would be profound. Without constant economic growth, the trillions of dollars in “safe” investment assets won’t seem so safe anymore. The capitalists, who depend on regular return rates, will be in trouble. The assumption that everything will ultimately rise is one of the unspoken pillars of capitalism. Yes, capitalism will continue, and yes people will continue to get rich, but they won’t do so automatically.

This doesn’t mean wealth disparity will vanish, though. If anything, it could concentrate wealth into fewer hands, into people like Warren Buffet whose investment skills transcend the universal macroeconomic bull.

For governments, consumption could become a national crisis. On Singles Day, Alibaba sold a record $14 billion worth of goods in China. Whereas the U.S. had its Thanksgiving Parade with High School marching bands and morning talk show hosts, China had Daniel Craig and Kevin Spacey reprise their roles in James Bond and House of Cards with a backdrop of Olympic fireworks and dancing. Singles Day makes Black Friday (the day after Thanksgiving in the U.S.) look like a bake sale. But this is all a leading indicator that governments are desperately trying to pump up demand. China has become too expensive of a place to manufacture and export goods, so they needed that spending spree to sustain their economy via domestic demand.

This call to consumption is similar to how after 9/11, George W. Bush asked Americans to stick it to the terrorists by shopping. Maybe some day, when the Treasury has exhausted all its economic tricks, and the economy doesn’t grow even during a supposedly favorable business cycle, the government will turn again towards consumption as a national emergency.

Either way, as Internet pioneer Marc Andressen said in his tweetstorm on secular stagnation, the people might ultimately not care:

In this world, we can have massive advances in real standards of living even w/formally low investment, GDP, & productivity growth.”
Beyond that, a world where 7 billion people decide they really do want and deserve an upper-middle-class Amercan-equiv lifestyle…

In other words, when everybody can max out their consumption with an upper-middle-class lifestyle, the upper-class will be the only ones complaining about GDP growth.