28/8/2012: Debt- v Equity-led Funding and Systemic Crises

Apparently, there's been some serious movements in today's banks CDS, signaling some pressure building up in the system and potentially a disconnect between equity markets and bond markets. This wouldn't be the first time the two are mis-firing in an almost random fashion. In the longer-term, however, such episodes are very troubling for a good reason - long term imbalances build up in the two sources of capital funding is hard to unwind. It turns out, however, the difficulty of unwinding these is non-symmetric.

Last week's NBER Working Paper number 18329 (link here), titled "Debt- and Equity-led Capital Flow Episodes" by Kristin J. Forbes and Francis E. Warnock looked at "the episodes of extreme capital flow movements—surges, stops, flight, and retrenchment... [leading to] the question of":

• Which types of capital flows are driving the episodes and 
• If debt- ( bonds and banking flows) and equity-led (portfolio equity and FDI) episodes differ in material ways. 

"After identifying debt- and equity-led episodes, we find that most episodes of extreme capital flow movements around the world are debt-led and the factors associated with debt-led episodes are similar to the factors behind episodes identified with aggregate capital flow data. In contrast, equity-led episodes are less frequent, more idiosyncratic, and differ in nature from other episodes."

The study uses data on 50 emerging and developed countries starting with 1980 (at the earliest) and running through 2009.

The study found that "the vast majority of extreme capital flow episodes across our sample—80% 

of inflow episodes (surges and stops) and 70% of outflow episodes (flight and retrenchments)—are 
fueled by debt, not equity, flows."

After that, the paper develops analysis of "the factors that are associated with debt- and equity-led episodes of extreme capital flows. We follow the Forbes and Warnock (2012) analysis here by describing capital flow episodes as being driven by specific global factors, contagion, 

and/or domestic factors." 

The study found that: "to a first approximation equity-led episodes appear to be idiosyncratic, bearing 
little systematic relation to our explanatory variables. Notably, even the risk measures that were 
highlighted in Forbes and Warnock (2012) as being significantly related to extreme movements in 
aggregate capital flows have little or no significant relationship with equity-led episodes. In contrast, 
risk measures are important in explaining debt-led episodes; when risk aversion is high, debt-led surges 
are less likely and debt-led stops are more likely. Contagion, especially regional, is also important for 
debt-led episodes. Country-level variables are largely insignificant, except for domestic growth shocks; 
debt-led stops are more likely in countries experiencing a negative growth shock and debt-led surges are more likely in countries with a positive growth shock."

Perhaps in a warning to the policymakers currently embarking on financial repression path for dealing with the ongoing crises, "capital controls have little or no significance in  both equity-led and debt-led episodes, as also found in Forbes and Warnock (2012)."

Of course, we have to keep in mind that the current crisis is really a debt-led capital markets crisis, both at the corporate and sovereigns levels. And it is symmetric both for the US and Europe, where the main difference is not as much in equity vs debt financing, but in intermediated vs direct debt issuance.