The antidote to post debate depression: facts and issues!
I am psychologically unfit for productive work today after the depressing farce of a “debate” last night. I was thus extra excited to see this Vox article from Tim Lee providing an excellent summary of the major storylines in our economy in 27 charts.
What better way to counter the existential doubt, dreams of a dystopian future, and vapid bickering ringing in our ears than to dig into some meaty charts and facts and plot a better course for our country’s future together?
As many of you know, I spent my summer immersed in the major issues of the modem economy in preparation for a book that ended up not coming together. From my reading, Lee captures most of the salient issues. The economy is in a state of transition. Growth is slowing but we don’t really know why; and yet we also aren’t sure if it’s unusual for growth to be slow, or if the preceding postwar period was just exotically high, possibly because of the unusual circumstances of two world wars and a Depression. I’m hoping my additional commentary below starts a dialogue around these issues.
The below are my “live blog” reactions tapped out on my phone as I read through the charts. The numbers correspond to the charts in Lee’s article. Charts are taken from the article.
2. Interesting that manufacturing employment as a number (rather than percent) didn’t shrink until recently -- it just stopped growing. Services have accounted for all of the growth in employment for the past forty or so years. This suggests some kind of equilibrium where for awhile manufacturing continued to add tasks even while automating others, but the rate of subtraction has over time eclipsed the rate of addition.
As we discover new things to do, these incremental services are often just monetizing activities that we used to do ourselves (eating out vs cooking, for example). We must let commerce into more and more aspects of our lives in order to keep people employed.
Relatedly, this is all really a story of Adam Smith and specialization. Each production task has been broken down to is constituent particles, including the delivery of the good, but also including our consumption. Outsourcing services from our lives allows us to focus on what is most important to each of us — we become more productive, more efficient. We no longer have to go to the grocery store. This can be amazing if used to extend our lifespans by increasing the percentage of present, emotional, high value interactions we have. If the time spent at the grocery store is instead spent telling stories as a family, we win. It can also put us in repetitive box, where we focus on the same few things to the detriment of our mental flexibility. This is the central tension of technology and specialization.
3. The flip side of creating more employment in the wake of productivity is that “new” activities are expensive. Humans, while productive, do not increase productivity over time without machines. But when you add machines to humans, often they become so productive that you don’t really need add many humans anymore.
This highlights the central tension of economics and politics: how to maintain the integrity of our institutions while also increasing our quality of life. Quality of life comes from making goods cheaper and more accessible through higher productivity (primarily driven by automation and trade).
The faster that the economy moves upward, the more disjointed it becomes from the human lifespan. Growth policies that encourage and fund innovation can speed industrial development; incumbent policies that privilege the current winners entrench the system and retard progress. If industries birth and die faster than humans, then humans trained in those industries get left behind each generation (See: extensive discussion in second Presidential debate about coal). This quite fairly enrages those stuck in the tumult, and turns them against growth policies and towards incumbent policies.
Since we all have an interest in those growth policies, we all have an interest in making sure that those who are cast aside by them are helped back onto their feet. And we have an interest in creating a system where such aid is promised ahead of time, to encourage rush taking and acquiescence in the face of progress.
Thus politics becomes the ballast of economics, working hand in hand to sail faster while holding the ship upright.
Long term trends
4. Low interest rates are a source of stability. Reading Hamilton, his main objective as first Treasury Secretary was to set the nation on a long run course of low interest rates, which would provide financial flexibility (in his thinking, to allow for financing of wars to defend the central government). We can see how successful he was in this chart.
This chart also highlights how the "problems" we view today are all in comparison with the runaway growth of the postwar years, which we know were unusual for many reasons, including (as Picketty notes) much of the wealth base had been destroyed in the preceding fifty years and more equality reigned. With these conditions, growth surged add the world converged back to where should be, and gains were distributed because more people started near the same starting line.
Is that period really a fair comparison historically? On the other hand, is acknowledging the difficulty of this comparison abandoning hope of high growth in the future? Do we abandon hope of high growth in the developed world, but retain it for those countries who can still converge to our standard of living? Or do the overall conditions signify that everyone is reverting to a hierarchical, slow growth world?
If we like the growth of the postwar period, then we should probably also strive to emulate the relative equality and high public and private investment in basic and applied research that were it’s essential ingredients.
7. The lack of inflation combined with the preceding output gap is the origin of much controversy today. These two facts signal that there is unused capacity waiting to be unlocked -- if there were not, then more money would be causing inflation. On the other hand, the output gap does not seem to be closing.
One theory is that long term unemployment damaged our potential output by forcing too many workers out of the labor force; more money is bringing them off the sidelines. I don't think the scale of workers lost (5-10M at most) is enough to account for the output gap -- rather I think the gap reflects an underinvestment in capital equipment to replace more workers. Perhaps this will clarify soon, as workers have started to rejoin the labor force over the past few months.
For reference, this is one of the main determinants of whether the Fed should raise interest rates. If you believe that there is more slack to absorb additional funds, then you should keep rates low; if on the other hand you think we are at full capacity and inflation is just around the corner, you should raise rates.
Robots and automation
9. The robots chart is misleading. I would venture that most job loses come from software -- Quick Books replacing accountants, for example. Software does not count as robots, but this is a meaningless semantic difference. (Edit: Tim Lee notes correctly that robots will matter more in some industries than others. Further, that some combination of robots and software will be needed to really impact industries like health care and retail. I agree with this view. We are I think in agreement that we’d want to look at both robots and software before making judgments on impact of tech, however).
10. This is the chart that got me really engaged in this work. To start with, Colin makes a good point that part of this is because health care costs have risen, so these charts do not show the additional burden of health insurance paid to insurance companies, not employees. That doesn’t likely close the whole gap though.
The separation of productivity from labor more likely signifies a decreased relative importance of human labor compared to machine labor, combined with (as Lee notes) increased inequality.
Inequality acts on this chart because the “new” jobs we have invented to deliver goods to customers are fundamentally low productivity growth jobs (how many heads cam a barber cut per day? How many people can a Walmart greeter greet?). So some people are in an economy of high productivity, collaborating with machines to drive higher quality of life. And others are relegated to a world of low productivity, because a) those are the jobs we have and b) many people, particularly in an older generation, were not trained to work with computers, and finally c) many people were stranded in rural towns when factories closed, and are unwilling or unable to uproot their families from the countryside and move to increasingly expensive and culturally foreign big cities for work.
Bonus chart: We can see this in unprecedented displacement in the Beveridge curve which measures open jobs relative to the unemployed. This recession has been marked by lots of unemployed people, but also lots of open jobs, highlighting a potential mismatch between what the economy needs and what workers have / where they are.
We solve these problems with training, welfare, and heavy relocation incentives. We also need to find ways to make service kinda more productive to grow wages, but also create even more service jobs to shelter the workers displaced by that productivity. It is an open question whether there are enough additional jobs. The low wages currently signal that there is not a lot of competition for labor, but whether that is a result of a natural end to employment or some structural factor we are getting wrong is still up in the air.
But we also recognize that we are not going to totally solve it at this late stage, and we resolve for the future to be proactive, never to abandon a generation to the whims of technology. We do this because it is ethical, but also because it makes the system more sustainable: we don’t want to dance with another Trump every election.
Politics and the future
25. This is the scariest chart of them all. There are some rich families with savings, but then everyone else has effectively nothing saved for retirement. Is it a surprise that our politics is being turned into angry backlash at cultural change throughout the world? Note the difference in Brexit voting between old and young, and the Bernie Sanders movement in the U.S.
There is an inter generational conflict occurring because the current older generation grew up in the exceptional postwar period, and circumstances have changed. Anti communism is no longer the most important Rorschach test, and growth is not resuming. If you predicated your future on continued good times, and financed purchased with debt, and you see the good times ending... It is not impossible to understand why anything that has changed (more foreigners, politicians, more liberal social values) attracts blame. The problem is that these things are an effect, not a cause, of the structural changes that have transpired.
27. The decline in startups is fascinating. Of course not all startups are future billion dollar ventures, but there is evidence (Haltiwanger 2011 if I remember correctly) that those types of startups are actually decreasing as well (the evidence is that the skew of startup success over time is becoming more narrow, e.g. there are fewer breakaway victories). At the same time, incumbent businesses are becoming more resilient to disruption, either through adaptation or acquisition. Incumbents employ a greater share of the population over time. Computers have allowed them to consolidate and homogenize markets, which is why every town in America looks the same today.
So long as the adaptation occurs, this is theoretically okay for the economy; but a decline in new challengers might signal over the long haul a decline in the need for large incumbents to react.
Going back to Hamilton, part of the goal of low interest rates was to encourage entrepreneurs. We should be thinking about whether our current system of limited IPO, concentrated venture capital, and unbridled large incumbents provides the right dynamism for the future.
I’d love to start a discussion about any of this in the comments or in real life. What do you think? Write below or give me a call. If Hillary and Donald won’t talk about it, at least we can do our part.