Do Banks Pass Through Credit Expansions to Consumers Who Want to - PowerPoint PPT Presentation

Do Banks Pass Through Credit Expansions to Consumers Who Want to Borrow? by Agarwal, Chomsisengphet, Mahoney, Stroebel Discussion by Christopher Carroll 1 with help from David Low, 2 Scott Nelson, 2 , 3 and Jialan Wang 2 1 Johns Hopkins University ccarroll@llorracc.org 2 Consumer Financial Protection Bureau 3 MIT NBER EFG Meeting, San Francisco, 2016-02-19

Real Question:

Real Question: Did the Banks Kill the Consumer? Retail Sales

Would Anyone Be So Bold?

Would Anyone Be So Bold? J’accuse!

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011)

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011)

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011)

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011) Problem: No serious attempt at quantification of effect of changes in credit supply ‘ S ’ on C

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011) Problem: No serious attempt at quantification of effect of changes in credit supply ‘ S ’ on C Paper’s motivating relation to this Q: ◮ Assume ‘ S ’ relates to measurable credit market conditions

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011) Problem: No serious attempt at quantification of effect of changes in credit supply ‘ S ’ on C Paper’s motivating relation to this Q: ◮ Assume ‘ S ’ relates to measurable credit market conditions ◮ Not, say, ‘animal spirits’ or ‘panic’

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011) Problem: No serious attempt at quantification of effect of changes in credit supply ‘ S ’ on C Paper’s motivating relation to this Q: ◮ Assume ‘ S ’ relates to measurable credit market conditions ◮ Not, say, ‘animal spirits’ or ‘panic’ ◮ When Fed/Treasury repair S to banks, do banks pass it on?

Would Anyone Be So Bold? J’accuse! ◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011) Problem: No serious attempt at quantification of effect of changes in credit supply ‘ S ’ on C Paper’s motivating relation to this Q: ◮ Assume ‘ S ’ relates to measurable credit market conditions ◮ Not, say, ‘animal spirits’ or ‘panic’ ◮ When Fed/Treasury repair S to banks, do banks pass it on? ◮ And, if so, do consumers consume?

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth ◮ a la Modigliani (1966), and FRBUS

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth ◮ a la Modigliani (1966), and FRBUS 2. Increase in Uncertainty

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth ◮ a la Modigliani (1966), and FRBUS 2. Increase in Uncertainty ◮ Carroll, Slacalek, and Sommer (2012)

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth ◮ a la Modigliani (1966), and FRBUS 2. Increase in Uncertainty ◮ Carroll, Slacalek, and Sommer (2012) 3. Change in Growth Expectations

There are Other Suspects in the Murder Mystery ... 1. Collapse in Household Wealth ◮ a la Modigliani (1966), and FRBUS 2. Increase in Uncertainty ◮ Carroll, Slacalek, and Sommer (2012) 3. Change in Growth Expectations ◮ Nobody (yet)

Authors’ Great Data on Originations

Authors’ Great Data on Originations are a Small Sliver ... Source: CFPB Consumer Credit Panel

... of a Small Sliver of Consumer Credit

Sometimes You do Find Your Keys Under Lamppost ... Their lamppost brightly illuminates this specific question: During the period of unprecedented economic turmoil covered by our dataset, did banks pass through credit expansions to consumers who wanted to borrow by quickly adjusting the availability of credit through the channel of the flow of offers for new credit cards? Their answer: No.

Sometimes You do Find Your Keys Under Lamppost ... Their lamppost brightly illuminates this specific question: During the period of unprecedented economic turmoil covered by our dataset, did banks pass through credit expansions to consumers who wanted to borrow by quickly adjusting the availability of credit through the channel of the flow of offers for new credit cards? Their answer: No. 1. Mark Twain: “A bank is an entity that will lend to you just when you don’t need it.”

Sometimes You do Find Your Keys Under Lamppost ... Their lamppost brightly illuminates this specific question: During the period of unprecedented economic turmoil covered by our dataset, did banks pass through credit expansions to consumers who wanted to borrow by quickly adjusting the availability of credit through the channel of the flow of offers for new credit cards? Their answer: No. 1. Mark Twain: “A bank is an entity that will lend to you just when you don’t need it.” 2. Their modification: “Any extra credit offers went mostly to people who didn’t want to spend.”

Here’s Another Lamppost Illuminating this Sliver 800 600 # of mail pieces (MM) 400 200 0 2000m1 2002m1 2004m1 2006m1 2008m1 2010m1 2012m1 Month Mintel Data on Credit Card Offers (A Fairly Direct Measure of “Supply”) Tells Much the Same Story

The Analysis that Leads to this Conclusion is ... ◮ Meticulous

The Analysis that Leads to this Conclusion is ... ◮ Meticulous ◮ Clever

The Analysis that Leads to this Conclusion is ... ◮ Meticulous ◮ Clever ◮ Convincing

The Analysis that Leads to this Conclusion is ... ◮ Meticulous ◮ Clever ◮ Convincing ◮ Impressive (743 RD’s)

... But Not Obviously Generalizable Given the vast literature on the peculiar nature of competition in credit card markets ◮ Ausubel (1991) through Grodzicki (2015) Of course, it is possible that very similar mechanisms operate in the other 99 percent of the credit market. But that remains to be proven.

I Totally Buy Their Description of the Trees 1. Different kinds of consumers behave very differently

I Totally Buy Their Description of the Trees 1. Different kinds of consumers behave very differently 2. Credit offers carefully tailored by type to max profits

I Totally Buy Their Description of the Trees 1. Different kinds of consumers behave very differently 2. Credit offers carefully tailored by type to max profits

I Totally Buy Their Description of the Trees 1. Different kinds of consumers behave very differently 2. Credit offers carefully tailored by type to max profits Confirms that competition in this market is all about adverse selection and moral hazard

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? Problem: Is this small or large?

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large?

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large? ◮ sounds small

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large? ◮ sounds small ◮ But, perfect competition would say such profits should be zero

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large? ◮ sounds small ◮ But, perfect competition would say such profits should be zero

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large? ◮ sounds small ◮ But, perfect competition would say such profits should be zero Alternative interpretation:

On Reflection, Not Quite As Sure About the Forest ACMS estimate what they say are “small” effects on profits from ∆ S ◮ By how much could banks raise profits by changing limits? ◮ By hiring ACMS! Problem: Is this small or large? ◮ sounds small ◮ But, perfect competition would say such profits should be zero Alternative interpretation: ◮ OMG! Banks were actually profitable in this business line