Bank Lending During the Financial Crisis of 2008

Victoria IvashinaDavid Scharfstein

Harvard Business School

To understand the spill over of the crisis from financial sector to real sector through the lending channel

– Did bank lending fall?

– If so, was it a contraction in demand or supply?

• How did different types of banks respond to the crisis?

Goal

Prior:

Source: Federal Reserve Board, Assets and Liabilities of Commercial Banks in the United States, (http://www.federalreserve.gov/releases/h8). Not seasonally adjusted, adjusted for mergers.

C&I Loans by Domestically Chartered Commercial Banks

Data

• Reuters DealScan: Origination of large loans (primarily syndicated loans)Self reported data:

-advertise-reflect market conditions-most importantly, receive league tables credit (published quarterly)

• Data through December 31, 2008• US companies

• Primarily US banks but also includes domestic affiliates of foreign banks

• From Aug ’08 to Oct ’08, top three US banks Citi, JPM, BAC originated 62% of the loans to the US companies, followed by Morgan Stanley with 4% of the loan origination

Basic Facts: Bank Lending Falls

Total Loan Issuance, US Corporate Loans (Amount and Number of Loans)

• New lending in 2008 was significantly below new lending in 2007, even before the peak period of the financial crisis

• The decline in new loans accelerated during the financial crisis, falling by 47% in dollar volume and 33% in number of issues in 4th quarter of 2008 relative to the previous quarter (79% and 61% with respect to the peak)

Basic Facts: Bank Lending Falls

Total Loan Issuance, US Corporate Loans (Billion USD)

0

100

200

300

400

500

600

700

800

QI '

00Q

II '0

0Q

III '

00Q

IV '0

0Q

I '01

QII

'01

QII

I '01

QIV

'01

QI '

02Q

II '0

2Q

III '

02Q

IV '0

2Q

I '03

QII

'03

QII

I '03

QIV

'03

QI '

04Q

II '0

4Q

III '

04Q

IV '0

4Q

I '05

QII

'05

QII

I '05

QIV

'05

QI '

06Q

II '0

6Q

III '

06Q

IV '0

6Q

I '07

QII

'07

QII

I '07

QIV

'07

QI '

08Q

II '0

8Q

III '

08Q

IV '0

8

Look at the loan issuance across three categories:

• Restructuring loans (M&A, LBOs, and stock repurchases) vs. Real investment loans (working capital or general corporate purposes)

• Non-investment grade vs. investment grade loans

• Term loans vs. revolving lines

*Based on 2007 numbers

Real investment loans

Restructuring loans

Corp. purposes24%

Work. Capital7%

Recap.5%LBO/M&A

56%

Real estate3%

CP backup2%

Exit financing2%

Proj. finance1% Debtor-in-poss.

0%Other0%

Restructuring loans vs. Real investment loans:

Restructuring loans vs. Real investment loans:

• There is a drop in both real investment loans (working capital or general corporate purposes) and restructuring loans (those for M&A, LBOs, and stock repurchases)

0

50

100

150

200

250

300

350

400

450

QI '07 QII '07 QIII '07 QIV '07 QI '08 QII '08 QIII '08 QIV '08

Restructuring Loans Real Investment Loans

Non-investment grade vs. investment grade loans:

• In the last quarter of 2008, non-investment grade loans fell by 82% relative to the prior quarter, while investment grade loans fell by 77%

0

50

100

150

200

250

300

350

400

QI '07 QII '07 QIII '07 QIV '07 QI '08 QII '08 QIII '08 QIV '08

Non-Investment Grade Investment Grade

Term loans vs. revolving lines:

• Sep–Nov 2008, revolving credit facilities and term loans both declined, but the decline in term loans (67%) was larger than the decline in revolving facilities (27%)

0

50

100

150

200

250

300

350

400

QI '07 QII '07 QIII '07 QIV '07 QI '08 QII '08 QIII '08 QIV '08

Term Loans Revolving Lines

Is drop in lending a supply shock or demand shock?

Supply effect if bank characteristics affect lending

• Effect of deposit base on lending– Banks that are more reliant on short-term debt have

difficulty rolling over debt and will have to cut lending more– Thus, banks with strong deposit base will cut lending less

Caveat: insured deposits

C&I Loans by Domestically Chartered Commercial Banks

Source: Federal Reserve Board, Assets and Liabilities of Commercial Banks in the United States, (http://www.federalreserve.gov/releases/h8). Not seasonally adjusted, adjusted for mergers.

Reconciling the facts:

• Borrowers draw down their credit facilities:

OutstandingLoanst

= OutstandingLoanst-1 + NewLoanst +

Drawdownst –LoanRetirementst

Borrowers draw down their credit facilities

34 firms, nearly $27

billion just in this sample (i.e., 26% of

the jump)

Borrowers draw down their credit facilities for precautionary reasons

“ Drawing down these funds is a prudent liquidity measure. Ensuring access to our liquidity to the fullest extent possible at a time of ambiguity in the capital markets is in the best interest of our customers, suppliers, shareholders, and employees.”Dana Corp. explaining $200 mm drawdown.

“ In light of the uncertain market environment, we have made this proactive financial decision to increase our liquidity and cash position and to bridge our access to the debt capital markets.”Duke Energy explaining $1 bn drawdown.

“ The Company believes that these actions were necessary to preserve its availability to capital due to Lehman Brothers’ level of participation in the Company’s debt facilities and the uncertainty surrounding both that firm and the financial markets in general.”FairPoint Communications explaining $200 mm drawdown.

Source: SEC filings

Is drop in lending a supply shock or demand shock?

Supply effect if bank characteristics affect lending

• Effect of deposit base on lending– Banks that are more reliant on short-term debt have

difficulty rolling over debt and will have to cut lending more– Thus, banks with strong deposit base will cut lending less

Caveat: insured deposits

• Effect of revolving line exposure on lending– Banks with large exposure to revolving lines will cut new

lending more But, note that banks with large exposure to revolving lines alsotend to have a strong deposit base (Kashyap, Rajan, and Stein, 2002)

Revolvers and Deposits

y = 0.34+ 0.45xR² = 0.47

0

0.2

0.4

0.6

0.8

1

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Rev

olve

rs/T

otal

Loa

ns

Deposits/Assets

Lehman exposure

Example: Tribune Co. 750 $MM revolving line

JPM (375 $MM)

Tribune needs/draws 300 $MM150 $MM

150 $MM

With Lehman out of the picture:

LehmanTribune needs/draws 300 $MM

300 $MM

Lehman (375 $MM)

JPM (375 $MM)

Empirical Approach

Define three windows:Pre-Crisis: August 2006 – July 2007 Crisis I: August 2007 – July 2008Crisis II: August 2008 –December 2008

Dependent variable: %Δ Total number of loans = [Mean(#loans per month)Crisis II / Mean(#loans per month)Base – 1]

where base = Pre-Crisis or Crisis I(Median is -55% relative to Pre-crisis, and -39% relative to Crisis I)

%Δ Total volume of loans per month (defined analogously)

Regression: %Δ Total number of loans on lagged Deposits/Assets

Results

Results

Economic magnitude: banks with revolving line exposure to Lehman one standard deviation above the mean (12%) cut lending by 44%, while banks with Lehman exposure one standard deviation below the mean (0%) cut lending by only 25%

Robustness: Revolving lines vs. term loans

Results

Robustness: CDS

Implications

• Measurement issues• Financial crisis had an adverse effect on supply of credit at

the bank level

• However, it is possible that there was a drop in loan demand, which was satisfied by the banking sector as a whole though a shift in supply from low deposit banks to high deposit banks

– Raises an important question: Can firms easily switch from one bank to another? If not, there will be real effects of the financial crisis even if there has been a drop in demand