The financial sector in the United States has grown from being about twice the size of GDP in 1980 to more than five times today (see Figure 1). The debate in the macroeconomics and finance literature on what caused this stark trend is still open. In my job market paper, I claim that the joint occurrence of higher inequality and lack of public safe assets can be seen as a major determinant.

Figure 1. Financial assets intermediated as a share of GDP and top 1 percent income share in the United States over the long run.

Understanding the mechanisms: Macro meets Finance

To understand the reason and the mechanism, it is worth unpacking several forces at play. In the paper, I show that the amount of savings redeployed towards financial markets increased as a result of higher inequality. However, when investors have a higher amount of assets under management, they also need to solve a portfolio allocation problem on how to diversify and allocate such savings across different asset classes: for instance, risky vs. safe assets. That is where macro meets finance. The paper highlights that there are natural reasons for investors to hold both risky and safe asset classes as we see in the data: a property that I define as “endogenous portfolio complementarity”.

Yet, what happens when there are simply not enough safe assets produced by the government (say, in the form of treasuries)? After all, the creation of public safe assets responds to rationales that are different from creating an elastic supply for the investors. By simple demand and supply logics, one can predict that the price of the safe assets will rise as demand surges. But higher prices mean lower interest rates, and lower interest rates mean lower debt issuance costs. In other words: issuing debt becomes cheaper because the interest rates to borrow decrease. It follows that households have incentives to finance a larger fraction of their consumption through debt.

These circumstances create ideal conditions for financial intermediaries to step in and produce private safe assets by transforming the higher debt that households are now willing to issue into quasi-safe assets that investors are willing to hold for precautionary motives (something akin to hedging your bets and insuring your portfolio).

In this respect, the paper shows that most of the growth of finance has been driven by other non-bank financial institutions – what usually goes under the name of shadow banking sector. Therefore, the changes in size and composition should really be thought of as two sides of the same coin.

Effectively, the shadow banking sector stepped in to make financial markets become more complete and better at insuring at the micro level. However, at the macro level this happened by using higher amounts of debt in absolute terms and as a fraction of households’ income – something picked up the leverage measure (which can also be a symptom of riskiness). Figure 2 helps visualizing the model mechanics.

Figure 2. Model diagram.

Feedback effects

To make matters more interesting both from a theoretical and policy perspective, a larger demand for financial securities increases not only quantities but also prices; however, asset prices are a mechanism which exacerbates inequality itself! To put it differently, more savings deployed in financial markets increase quantities and prices of the financial assets demanded, but higher assets valuations make their holders wealthier which induces even more savings to be reinvested – a mechanism that feeds back on itself. This type of theoretical prediction is also substantiated by data and has large policy implications. Households finance has argued for a long time about the importance of allowing larger swaths of the population to hold financial assets. This paper confirms that and adds on the potentially very large welfare implications also from a macroeconomic standpoint. More work on this is certainly needed in the future.

Running a few policy experiments

The paper is overarching enough to be structured as a quantitative exercise where the model is brought to data to test the magnitudes of the different forces. To keep the feedback effect distinct from the direct channel, I start with an exogenous primitive change such as the decline of the labor share. This is a very conservative position given that inequality has been driven by many other secular factors. Changing the labor share, however, has the benefit of maintaining the model very parsimonious and easy to inspect (more ingredients can always be added). The model predicts that the decline of the labor share alone leads to an increase of the top 10 percent share of the distribution (the one holding most of the financial assets) by over 10 percent, which can explain up to 25 percent of the rise of the shadow banking sector. Adding more sources of inequality can make the results even stronger.

With that in hand, I run a few policy experiments to see how the variables react. In particular, I impose a dividend tax, whose proceeds are rebated to households as a subsidy. I run a similar experiment by using the higher fiscal capacity to back more public safe assets (rather than rebating it to households). And finally, I run a macroprudential policy in the form of a Tobin tax, where I directly tax the shadow banking holdings, and then use the proceeds to expand the amount of public bonds. The strongest results are by far the ones emerging from the last experiment. A measure that directly targets shadow banking holdings in the form of a 4 percent tax is able to curb them to pre-1980 level, which is remarkable considering the relatively small size of the tax.

Broader implications: Why does it matter?

A large financial sector can have overarching implications for the overall economy. We saw in 1929 and 2008 the effects that a financial crisis has when the financial sector is particularly sizable. The larger the sector, the larger and more pervasive the repercussions that can come from it when financial market instability strikes. That claim becomes even more salient if the growth is determined by actors outside the traditional regulatory perimeter (shadow banking sector), which is harder to scrutinize and supervise. Not to mention that a financial sector that grows larger intrinsically through higher debt outstanding and leverage makes growth and crises (i.e. trends and cycles) intertwined at their core. This has implications for the vast world of financial regulation and macroprudential policies.

In addition, there is vast empirical literature that does not find a positive effect of finance on economic growth and investments for advanced economies after a certain threshold is passed (Arcand, Berkes, Panizza, 2015). Therefore, the welfare implications of a large financial sector are unclear even when no crises emerge. Understanding what leads to such size then becomes salient.

Finally, to the extent that only a small subsection of the population owns financial assets (the top 10 percent), when financial assets valuations increase in good times, this has large distributional consequences that exacerbate inequality. Needless to say, the welfare effects of such distributional impacts deserve special attention.

Overall, my job market paper tries to build a bridge across several strands of the macroeconomics and finance literature to provide an answer to previously posed questions (Philippon, 2015, above all). However, as is often the case, even more interesting questions emerge as a result of this analytic work.

About the Author

Dario Laudati is a Postdoctoral Research Scholar at Columbia University. He holds a PhD in Economics from the University of Southern California

He is interested in macroeconomics and finance, and he is currently on the economics and finance job market. To learn more about his research, visit: https://dariolaudati.com

Social Media Handles:

• LinkedIn
• X
• Bluesky
• Mail