SLIDE 1 Do Banks Pass Through Credit Expansions to Consumers Who Want to Borrow?
by Agarwal, Chomsisengphet, Mahoney, Stroebel
Discussion by Christopher Carroll1 with help from David Low,2 Scott Nelson,2,3 and Jialan Wang2
1Johns Hopkins University
ccarroll@llorracc.org
2Consumer Financial Protection Bureau 3MIT
NBER EFG Meeting, San Francisco, 2016-02-19
SLIDE 2
Real Question:
SLIDE 3
Real Question: Did the Banks Kill the Consumer?
Retail Sales
SLIDE 4
Would Anyone Be So Bold?
SLIDE 5
Would Anyone Be So Bold?
J’accuse!
SLIDE 6
Would Anyone Be So Bold?
J’accuse!
◮ Eggertsson and Krugman (2011)
SLIDE 7
Would Anyone Be So Bold?
J’accuse!
◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011)
SLIDE 8
Would Anyone Be So Bold?
J’accuse!
◮ Eggertsson and Krugman (2011) ◮ Guerrieri and Lorenzoni (2011)
SLIDE 9
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
SLIDE 10
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
SLIDE 11 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’
SLIDE 12 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?
SLIDE 13 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?
SLIDE 14 There are Other Suspects in the Murder Mystery ...
- 1. Collapse in Household Wealth
SLIDE 15 There are Other Suspects in the Murder Mystery ...
- 1. Collapse in Household Wealth
◮ a la Modigliani (1966), and FRBUS
SLIDE 16 There are Other Suspects in the Murder Mystery ...
- 1. Collapse in Household Wealth
◮ a la Modigliani (1966), and FRBUS
- 2. Increase in Uncertainty
SLIDE 17 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)
SLIDE 18 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
SLIDE 19 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)
SLIDE 20
Authors’ Great Data on Originations
SLIDE 21
Authors’ Great Data on Originations are a Small Sliver ...
Source: CFPB Consumer Credit Panel
SLIDE 22
... of a Small Sliver of Consumer Credit
SLIDE 23
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.
SLIDE 24 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.”
SLIDE 25 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.”
SLIDE 26 Here’s Another Lamppost Illuminating this Sliver
200 400 600 800 # of mail pieces (MM) 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
SLIDE 27
The Analysis that Leads to this Conclusion is ...
◮ Meticulous
SLIDE 28
The Analysis that Leads to this Conclusion is ...
◮ Meticulous ◮ Clever
SLIDE 29
The Analysis that Leads to this Conclusion is ...
◮ Meticulous ◮ Clever ◮ Convincing
SLIDE 30
The Analysis that Leads to this Conclusion is ...
◮ Meticulous ◮ Clever ◮ Convincing ◮ Impressive (743 RD’s)
SLIDE 31
... 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.
SLIDE 32 I Totally Buy Their Description of the Trees
- 1. Different kinds of consumers behave very differently
SLIDE 33 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
SLIDE 34 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
SLIDE 35 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
SLIDE 36
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?
SLIDE 37 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?
SLIDE 38 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
SLIDE 39 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
SLIDE 40 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
SLIDE 41 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:
SLIDE 42 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
SLIDE 43 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 ◮ Proves they did indeed cut supply of credit ‘too much’
SLIDE 44 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 ◮ Proves they did indeed cut supply of credit ‘too much’ ◮ Only way to judge this is by comparison to “normal” profits
SLIDE 45 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 ◮ Proves they did indeed cut supply of credit ‘too much’ ◮ Only way to judge this is by comparison to “normal” profits ◮ Unfortunately, data start in ‘abnormal’ 2008q2
SLIDE 46 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 ◮ Proves they did indeed cut supply of credit ‘too much’ ◮ Only way to judge this is by comparison to “normal” profits ◮ Unfortunately, data start in ‘abnormal’ 2008q2 ◮ And we’re not yet back to ‘normal’
SLIDE 47
Intermediate Results More Exciting than Conclusion!
SLIDE 48
Intermediate Results More Exciting than Conclusion!
Because helps us understand mechanisms important for HANK
SLIDE 49
Intermediate Results More Exciting than Conclusion!
Because helps us understand mechanisms important for HANK Estimates of the MPB by credit score
SLIDE 50
Intermediate Results More Exciting than Conclusion!
Because helps us understand mechanisms important for HANK Estimates of the MPB by credit score Proxy for wealth-to-income ratio
SLIDE 51
Intermediate Results More Exciting than Conclusion!
Because helps us understand mechanisms important for HANK Estimates of the MPB by credit score Proxy for wealth-to-income ratio Only other estimates I know of are preliminary results by Jialan Wang et al, whose experiment is quite different but results are similar (at least in showing big deviations from 0!)
SLIDE 52
Intermediate Results More Exciting than Conclusion!
Because helps us understand mechanisms important for HANK Estimates of the MPB by credit score Proxy for wealth-to-income ratio Only other estimates I know of are preliminary results by Jialan Wang et al, whose experiment is quite different but results are similar (at least in showing big deviations from 0!) Credit Score MPB-Wang et al MPB-ACMS V Low 0.58 Low 0.28 0.47 Med 0.31 0.43 High 0.19 0.23
SLIDE 53
Why Is This Exciting?
Such data could be used to answer the Whodunit. How? By looking at how different kinds of consumers behaved during the crisis. Setup: ‘Buffer stock’ model calibrated to micro data with
◮ 3 kinds of consumers: Low, Mid, and High credit score
SLIDE 54
Why Is This Exciting?
Such data could be used to answer the Whodunit. How? By looking at how different kinds of consumers behaved during the crisis. Setup: ‘Buffer stock’ model calibrated to micro data with
◮ 3 kinds of consumers: Low, Mid, and High credit score ◮ 3 hypotheses about shocks:
SLIDE 55 Why Is This Exciting?
Such data could be used to answer the Whodunit. How? By looking at how different kinds of consumers behaved during the crisis. Setup: ‘Buffer stock’ model calibrated to micro data with
◮ 3 kinds of consumers: Low, Mid, and High credit score ◮ 3 hypotheses about shocks:
◮ Wealth shocks (proportional)
SLIDE 56 Why Is This Exciting?
Such data could be used to answer the Whodunit. How? By looking at how different kinds of consumers behaved during the crisis. Setup: ‘Buffer stock’ model calibrated to micro data with
◮ 3 kinds of consumers: Low, Mid, and High credit score ◮ 3 hypotheses about shocks:
◮ Wealth shocks (proportional) ◮ Uncertainty shocks (increase in size of income shocks)
SLIDE 57 Why Is This Exciting?
Such data could be used to answer the Whodunit. How? By looking at how different kinds of consumers behaved during the crisis. Setup: ‘Buffer stock’ model calibrated to micro data with
◮ 3 kinds of consumers: Low, Mid, and High credit score ◮ 3 hypotheses about shocks:
◮ Wealth shocks (proportional) ◮ Uncertainty shocks (increase in size of income shocks) ◮ Credit availability (tightening of borrowing constraints)
SLIDE 58 Theoretical Response: Negative Shock to Credit
10 10 20 30 Months Since Shock 0.08 0.06 0.04 0.02 0.00 0.02 0.04 0.06 0.08 Median Log Consumption
Consumption Around Credit Supply Shock Constrained Middle Class Rich
SLIDE 59 Theoretical Response: Increase in Uncertainty
10 10 20 30 Months Since Shock 0.06 0.04 0.02 0.00 0.02 0.04 0.06 0.08 Median Log Consumption
Consumption Around Income Uncertainty Shock Constrained Middle Class Rich
SLIDE 60 Theoretical Response: Proportional Shocks to Wealth
10 10 20 30 Months Since Shock 0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Median Log Consumption
Consumption Around Wealth Shock Constrained Middle Class Rich
SLIDE 61
Very Cool Paper ...
◮ ... not persuasive in answering the Whodunit
SLIDE 62
Very Cool Paper ...
◮ ... not persuasive in answering the Whodunit ◮ But shows techniques and glimmers of tools that maybe can!
SLIDE 63 References I
Ausubel, Lawrence M (1991): “The failure of competition in the credit card market,” The American Economic Review, pp. 50–81. Carroll, Christopher D., Jiri Slacalek, and Martin Sommer (2012): “Dissecting Saving Dynamics: Measuring Wealth, Precautionary, and Credit Effects,” Manuscript, Johns Hopkins University, http://econ.jhu.edu/people/ccarroll/papers/cssUSSaving/. Eggertsson, Gauti B., and Paul Krugman (2011): “Debt, Deleveraging, and the Liquidity Trap: A Fisher-Minsky-Koo Approach,” Manuscript, NBER Summer Institute. Grodzicki, Daniel (2015): “The Evolution of Competition in the Credit Card Market,” The Pennsylvania State University. Guerrieri, Veronica, and Guido Lorenzoni (2011): “Credit Crises, Precautionary Savings and the Liquidity Trap,” Manuscript, MIT Department of Economics. Modigliani, Franco (1966): “The Life Cycle Hypothesis, the Demand for Wealth, and the Supply of Capital,” Social Research, 33, 160–217.
