A recent NBER working paper by Adam Taylor uses economic history to take a look at how an unprecedented level of private credit growth contributed to the severity of the financial crisis.
The title is “The Great Leveraging” because that’s exactly what’s happened. As measured by balance sheets and private loan levels, our banking system is larger (compared to GDP) than at any other point in history.
One of the most interesting insights of the paper is that despite the substantial growth of public debt in recent years, it is dwarfed by the banking sector.
“The scale of the increase in the balance sheets in the banking sector has effectively flipped the main credit risk nexus, measured by debt magnitude, from the sovereign side to the banking side.”
That transformation occured extremely rapidly beginning in the 1970s:
Photo: Adam M. Taylor
This is a widespread phenomenon. Every advanced country has a banking sector with a balance sheet in excess of its GDP.
Why the massive shift? Taylor points to a number of hypotheses. Banks may have changed their risk appetite after they rebuilt from the war, regulations were significantly relaxed since the 1970s, and the existence of the Fed as a lender of last resort in addition to the creation of deposit insurance may have introduced a degree of moral hazard into the system.
Another important factor that Taylor notes is capital flow between developed nations and rapidly growing emerging markets. While private capital has flowed to emerging markets, public capital going from emerging to developed markets dwarfs it.
In the aftermath of the Asian financial crisis, emerging economies have been hoarding reserves and safe assets at unprecedented levels. Those flows helped push rates ever lower.
All of these factors combined to create an enormous banking system that benefitted from seemingly never ending cheap capital. Banks funded projects that were not worthy, and the banking system grew ever larger and more fragile.
Photo: Wikimedia Commons
The important thing is that we learn from this crisis and take steps to prevent future ones, he writes.First, we need to account for private credit growth as a predictor of financial crises. Taylor’s forecast model of the likelihood of financial crises found that credit growth was a significant predictor of such events.
However, the groups tasked with maintaining financial stability did not monitor or substantially use indicators of credit growth. That’s something that needs to change.
The second is that public debt levels are a distraction when it comes to predicting financial crises.
When tested as an indicator of financial crises against Taylor’s credit model, public debt levels have not been a significant cause of such events over the past 140 years. Cases exist, like Greece, but they have been the exception rather than the rule. Crises like those in Ireland and Spain originated in the financial sector are much more common.
Additionally, credit driven booms followed by financial crisis result in worse recessions than normal. That contraction is magnified by the level of credit growth relative to GDP.
Where high public sector debts do matter is in the response to the crisis. According to Taylor, high credit growth in the run up to a financial crisis is associated with an additional 1-1.5 per cent drag on GDP. If you combine that with public sector debt around 100 per cent of GDP, that jumps to a nearly 4 per cent drag over the course of a 6 year recovery.
That historic insight couldn’t be more relevant to our current situation, in which the ability to stimulate the economy and avoid austerity has been held back by debt levels.
Taylor’s brand of analysis is a useful way to look at the crisis and the reasons our recovery has not taken off as we hoped.
For example, the knowledge that credit driven booms followed by financial crises lead to much deeper recessions should have informed the policy response in 2008. Policy is too rarely made with the long term in mind, especially in times of crisis. We can’t go back and avert this crisis, but we can avoid the next one.
Read the full paper here.