I very much liked this economic model linking inequality, wage stagnation, excessive leverage, expansion of the financial sector, and financial instability.

Here’s the breakdown:

–Factors outside the model lead to income stagnation for middle and low-income workers, while high-income households acquire more capital assets.   This increases the savings of wealthy households relative to lower-income households.

–In order to keep their living standards from declining, the middle class borrows more.  Financial innovations, including new types of securitization, increase the liquidity and lower the cost of loanable funds available to the borrowers.  As the authors put it:

The bottom group’s greater reliance on debt—and the top group’s increase in wealth—generated a higher demand for financial intermediation.

–The financial sector thus grows rapidly as do the debt-to-income ratios of the middle class relative to the wealthy.

–The combination of rising middle class debt and stagnant middle class incomes increases instability in financial markets, and the system eventually crashes.

Re this last part, I’d add that some unique US and later European developments (though not Canadian, interestingly—more highly regulated banking sector, ‘ey?) amplified the last step of the model.  The so-called innovations—things like securitization that increased the distance between loan originator and loan holder, which in turn led to loose underwriting standards—interacted in a particularly volatile way with deregulatory zeal and the belief that financial players would self-regulate.

The first way to evaluate a model like this is to ask if the predictions it generates show up in the data.  Their first two charts below show aggregate debt-to-income across all income groups grew consistently with the income share of the top 5% both before the Great Depression and Great Recession.  The third chart shows that, in fact, this increase was a considerably sharper in recent years for the bottom 95% than the top 5%.

None of this is dispositive—there are always other explanations.  It makes sense that strapped middle class families would leverage up more than flush rich ones, but that doesn’t imply an obvious role for inequality.  In that regard, a key causal linkage here is that inequality diverted wage growth from middle and lower income workers.  I’ve made that case in lots of places, but it’s particularly relevant in models like this that assign a causal role to inequality.

Also, while the size of the financial sector did grow relative to GDP over this period, it had been growing for decades, and they need to show it accelerated—there’s some acceleration over this period of growing inequality, but it’s not huge.

Still, their model rings true to me, and it’s compelling to think about the policy implications, as the authors do at the end of the paper.  They argue that greater bargaining power by workers would ensure a more equitable distribution of growth, putting downward pressure on the leverage steps in the model as described above.

I’m all for that, but given that unions represent 7% of the private workforce, and 12% of the total (the public sector, at 37%, is a lot more heavily unionized, but they’re under sharp attack), that’s a long-term solution at best.  In the meantime, I’d push full employment as another way of boosting the bargaining power of the middle class, though that too is obviously an awfully heavy policy lift these days.

Better oversight of the banking sector also pops out of the model—if the income dynamics lead to greater demand for loans, it makes sense to ensure that risk is not underpriced, as was the case during the housing bubble.  Same with tax incentives that incentivize leverage, like favored tax treatment for debt financing.

I’d add this model to the growing body of work, including my own, on the impact of greater inequality on the rest of the economy.  I’d also note recent work by Krugman, Gilens (the political dimension), and Acemoglu/Robinson (sustainable growth).

The importance of this line of thought is that it doesn’t simply point to high inequality as an outcome we don’t like for reasons of fairness or balance.  Instead, this work is looking at historically high levels of inequality as an input into models trying to figure out what’s gone wrong.   That strikes me as precisely the right way to approach the diagnosis.

Source: Kumhof and Ranciere, link above.