The central fact about the stock market is that you can’t explain short-term movements on the basis of any kind of objective factors, because prices are fundamentally self-referential. The short term return on stocks is dominated by capital gains, which means that when you’re deciding what a good price is today you’re just trying to guess what prices will be tomorrow.

Now, if you’re looking at longer-term changes, then at some point other factors come into play. With stocks, there are two broad considerations. One is the flow of profits that are ultimately going to accrue to the owners of those stocks, and the other is liquidity and credit conditions — when it’s difficult to get loans or to sell on short notice, you’re going to put a lower value on risky assets like stocks.

So if the expected flow of profit income going to stockholders declines, the value of stocks will decline. There are many reasons why these profit flows — and even more, people’s beliefs about them — may change. But one reason is if there’s a change in the share of output going to profits versus the share going to wages. You probably saw the headlines the other day about how concern over rising wages is leading some people to think the bull market is coming to an end. And that’s where you get into a more political story.

There is no question that there has been an uptick in wage growth in the past couple of years. I think over the most recent year average wages are up 2.6%, which by historical standards is pretty mediocre, but it’s significantly faster than we’ve seen for a number of years — really since before the Great Recession. Furthermore, this acceleration of wages is happening without even a hiccup of inflation and without any noticeable improvement in productivity. Which means it’s coming straight out of profits.

There’s a bit of a tightrope we have to walk in thinking about this. It’s important to emphasize that, even close to a decade after the official end of the last recession, this is not a full recovery in many ways, especially for people who work for a living. But it’s also true that the relative position of working people is noticeably better than it was a couple of years ago. If you look at the share of value-added going to labor in the corporate sector, it’s up to 59 percent today compared with 57 percent in 2010 to 2014. Now, two percentage points doesn’t sound huge, but that’s actually a significant shift of the pie from business owners to workers. It’s most of the way back from where we were before the financial crisis. It’s the biggest sustained rise in the labor share since the late 1990s.

So if you’re somebody whose income comes from the part of output that does not go to workers — which you are by definition, if you’re a shareholder — then you are not crazy to be concerned. There’s an important strand of Marxist economics on what’s called the cyclical profit squeeze. This argues that historically in the United States — and at least to some extent in other advanced capitalist countries — you’ll find that every recession is preceded by a period of declining profits, and that those declines are normally caused by exactly this: a shift in the share of output from capital to labor.

Some people have formalized this as a “Goodwin cycle.” This is a repeated process where you start from strong growth, low unemployment, a relatively favorable bargaining position for workers. That drives up wages at the expense of profits, until eventually at some point businesses refuse to invest because profits fall too far. The falling profit interrupts the reproduction of the system and you get a crisis where unemployment goes back up, and then that creates the conditions for restored profitability and renewed growth. Obviously the real world is not so cut and dried — the link from profits to investment is one questionable piece — but it’s a useful tool for thinking about our current situation.

On the other hand, there is absolutely no reason to expect an uptick in inflation. I think it’s clear that, at this point, talking about inflation as the reason to be concerned about wage increases is just a cover story. There is no evidence that wage growth passes through to inflation at all. Faster growth in wages, if it’s sustained long enough, may lead to faster productivity growth, which would allow profits to be maintained. Otherwise, we should expect a simple one-for-one tradeoff between wages and profits.

Again, we have to be able to keep two somewhat contradictory thoughts in our mind at the same time. On the one hand, the idea of secular stagnation — a more or less permanent tendency for spending to fall short of productive potential — certainly has not been proven wrong. This is still a relatively weak expansion by historical standards — the fraction of the population with a job, for instance, is still much lower than it was a decade ago. We’ve had years of historically low interest rates and large federal deficits, which conventional theory says should give a big boost to demand. Yet we are only now getting back to what’s at best an adequate level of growth.

But on the other hand, and this is just as important, we are also seeing that it is not the case that automation or globalization have permanently broken the link between aggregate spending and the economic position of workers. A sustained period of faster growth does increase the fraction of people with jobs, and it does improve the bargaining position of workers. It does allow people to demand higher wages than they were able to a couple of years ago. There was an idea a few years ago — among conservatives and libertarians especially, but also among some on the left — that we were entering a new period of structural unemployment where, because of robots or skills mismatch or whatever, millions of previously employed people were now unemployable. I think the experience of the past couple years gives strong evidence that this is not true.

When you have more money flowing through the economy, that’s going to mean more demand for labor, and that’s going to strengthen the bargaining position of people who sell their labor. There isn’t some sort of structural factor that has permanently rendered workers unable to claim a larger share of output in the form of wages.

Well there has been an upturn in business investment. In the corporate sector, at least, business investment, after being very weak for a number of years, is now near the high end of its historical range as a fraction of output.

I don’t think we should completely dismiss it, but it’s not the main story. The fundamental thing is that this is a system that’s unstable, but it’s unstable in both directions. A sustained period of growth, even moderate growth, is going to make lenders more confident, it’s going to make them relax standards, it’s going to make businesses more confident, more eager to expand. As you work through the backlog of excess capital goods that you had early in the recession, you are going to have to start investing again.

I do think the Fed could have made things worse. If you compare it to the European Central Bank, which took on the role of being an enforcer of austerity and using its leverage to roll back social protection and weaken labor rights and so on, we can appreciate the fact that the Fed has not done that — though you could say they didn’t have to, since that work has already been done in the United States. They were more willing to use the various tools at their disposal to try to soften the impact of the crisis and increase demand without any kind of quid pro quo.

You can definitely make a case that if they hadn’t done that, the crisis would have been worse. The problem is that after the immediate crisis period, the Fed has been trying to boost spending using a limited range of tools. For the most part they’ve operated through the peak of the financial system — by pushing down interest rates on bank reserves and government debt — in the hopes that this will eventually get passed through to the broader credit markets and to real activity. The major exception to this has been their willingness to directly support mortgage lending. I don’t think there’s much question that mortgage rates are lower, and mortgages more available, than if the Fed hadn’t been buying such a large fraction of mortgage debt over the past decade. Presumably this has made people more willing and able to buy houses, and also made homeowners more able to buy other stuff.

Interestingly, Bernanke says in his memoir that there was a serious discussion at the Fed about whether they should intervene to support the market for state and local government debt as well, to make it easier for distressed governments to keep borrowing. There’s debate about exactly how much legal space they have to do that, but they certainly have some and they did contemplate doing it. But ultimately they said no, that’s not our job, that’s for Congress.

I think that sort of intervention could have been much more effective than quantitative easing as it was actually carried out. If they’d been willing to be more aggressive, and in particular to shift the focus of their actions away from a few of levers at the center of the financial system to act more directly, closer to the real economy, closer to the ultimate borrowers, they could have had a larger impact on demand. Right now there seems to be a kind of smug “mission accomplished” feeling at the Fed. But I’m sure the next time there’s a serious downturn, conversations about using a broader array of tools are going to start up again.

I think the key here is leverage or debt. You don’t get a systematic unraveling of financial commitments from a fall in stock prices alone. You have to have enough units in the economy who have made legally binding commitments — debt contracts or something equivalent — so that if the price of some other asset falls, it means someone can’t service their debts, which means someone else can’t, and you end up with an unraveling of the whole network of commitments. That’s what happened in 2008.

Stock markets don’t create those sorts of commitments. Obviously there’s an expectation that businesses will deliver a profit to shareholders commensurate with the price of the stock, but there is no legally binding obligation to be met. And we don’t have a lot of stock purchases financed through borrowing; regulation has been pretty effective about that.

This is very different from housing, which is overwhelmingly debt-financed. Obviously, if you had another collapse in real estate prices, that would be a problem, but it doesn’t seem like we’ve had anything like the runup in housing prices that we had at the top of the boom ten years ago. On the other hand with something like bitcoin, there doesn’t seem to be any connection to the broader financial system. It’s not enough just to have an asset fall in value to create a financial crisis.

That is probably true to some extent. I think we have probably not paid enough attention to the importance of luxury spending by high-income households in maintaining demand.

There’s a widespread idea that basic Keynesian logic creates a common set of interests, because capital needs workers to be making adequate incomes to maintain demand. Now, even if you believe that, there’s still a contradiction, in that wages are obviously both a cost and a source of demand, and that always creates tension in terms of what path capitalists would like wages to follow. But it may be that you don’t even need wage growth as a source of demand, if rich people are willing to spend a high enough fraction of their income — or, if you have businesses investing enthusiastically, or rising government spending, or exports or some other source of demand.

Consumption spending by the rich is hard to nail down statistically; it’s harder to observe consumption than income. But to the extent that we can track it, it does seem that over the 2000s there was a large increase in consumption spending at the top of the income distribution. According to most of the research I’ve seen, consumption inequality rose by as much, if not more than, income inequality. That’s actually sort of surprising, because we normally think rich people spend a smaller fraction of their income, so the distribution of consumption should be more stable than the distribution of income.

I haven’t seen really recent numbers on this, but at least over the past fifteen or twenty years, the increase in income at the top seems to have been more or less matched by an increase in consumption at the top. So as long as that’s going on, a lack of demand from wage-earners isn’t a problem. The problem you do have is, as you say, that spending by wealthy households is more sensitive to the prices of stocks and other assets than the spending of ordinary people.

Yes, I did see that: a twenty-eight-hour work week!

I think that’s absolutely right. One way to look at it is that capitalists have a collective action problem. When conditions are good they want to sell everything they can produce. So they’re perfectly happy, individually, to pay somewhat higher wages or make other concessions, like a shorter workweek, to get the labor they need.

But for the capitalist class as a whole, they’re undermining their collective bargaining position as they all compete with each other over scarce labor. And so it’s the job of the central bank — or whoever the authorities are, but today it’s generally the central bank — to act in the collective interest of capitalists, to get them to stop undermining each other in their competition for workers by creating a condition of artificial capital scarcity, by reducing demand. In a way, you could say this is not so different from the collective action problem of climate change, except that this is one that capital and its political institutions take very seriously.

Another way of looking at it is that we have this class of people, capital owners, who notionally have a useful role in coordinating production. The problem is, the thing they are supposed to be providing — capital, in its various incarnations — isn’t necessarily scarce in today’s world. The more that the growth in production depends on science and technology, on large-scale coordination, on various social institutions — things that don’t fit the mold of private property — the less useful money-prices and markets are in coordinating it. The kind of means of production that you can buy with money are no longer the main limiting factor. And liquidity, the thing that financial markets are supposed to provide, is not scarce in any objective sense.

This is the situation of capital abundance that Keynes thought would be reached in a generation or two from when he was writing in the 1930s, which would lead to the euthanasia of the rentier. And he was right! The problem is, the rentier doesn’t want to be euthanized. Capital is not going to say, “okay, our work here is done, goodbye.” So to maintain the social position of money-owners, you have to create an artificial shortage of money, and that’s another way of looking at the job of the central bank.

The problem that you’re pointing to is, the system is not very robust to shortages of money. It’s not very robust to the kind of scarcity that has to be created to maintain the social position of capital. So if you are trying to manage it, you have to strike a delicate balance. If you focus too much on maintaining the bargaining position of capitalists, you interrupt the reproduction of the productive system, and you create situations where people aren’t able to meet their financial commitments. But on the other hand, if you’re too loose, as you said, you create a situation where capital is abundant and the bargaining power moves to workers, or perhaps to owners of land, or intellectual property, or other stuff that can’t be reproduced with money. It’s always a tightrope they have to walk.

The IG Metall contract is a good example of the danger on that side, from the point of view of the authorities. More broadly, there’s a strong argument that the low unemployment and sustained growth of the postwar period laid the foundation, not just for a stronger labor movement but for the whole range of social movements in that period. These developments are extremely complex and depend on the specific circumstances and conditions.

But it seems likely that the postwar period of sustained very strong growth — which of course was motivated by the need to compete with the Soviet Union and show the superiority of “our” system — made it possible not just to have an upsurge of workers’ movements, but the new social movements around gender, the environment, and so on. When you weaken the boss’s authority in the workplace that is going to undermine other hierarchies as well.

So while we wouldn’t want to be too mechanical about it, there are good historical reasons for thinking that movements to transform the system may actually be in a stronger position when the system is seemingly working well.