Comment on Adrian and McAndrews

John Taylor

Description

Comments on“The FdFederall R eserve’’s Primary Dealer Credit Facility” Tobias Adrian and James McAndrewsSGE Session on “The Fed’s New Lending Facilities”ASSA Meetings San Francisco JhJohn B. TaylorStanford UniversityJanuary 4, 2009Summary• Excellent paper on economic rationale for PDCF– Analogy: Broker‐Dealers (PDCF), Depository Institutions (PCF)• Focus onon role ofof the ReRepo market– Haircuts, margins, leverage, downward spirals (33:1 to 20:1)– Leverage first rose in recent crisis ((,SIVs, large rate cuts)• Structural changes over time:– Growth of broker‐dealers and hedge funds relative to dideposittory iitnstitituttiions– Trend toward tradable assets and liabilities– Repoo runs or margin spirals rarather than classic bank runsruns• Protections: Penalty rates, regulation, moral hazard• Main Comments:– Empirical Impact – Slippery slope to more facilitiesEmpiricall Findings• “credit default swap prices of pyprimary dealers fell after the introduction of the PDCF”• “To date, the PDCF seems to be having measurable effectcts on thethe repoo and otherother funding markets, as many primary dealers found that their perceived risk of default, as measured by credit default swap prices, declined after the facility’s introduction. ” • Yet nono evidence prpresentted in thethe paper, so lelet’s take a look at CDS rates for Merrill Lynch and Goldman Sachs (for example)…basis pointsmillions of dollars500400300150,000200125,000100100 ...

Informations

• Publié par: Ossi
• Nombre de lectures: 858
• Langue: English

Extrait

Comments on e e era eserve s Primary Dealer Credit Facility” Tobias Adrian and James McAndrews

SGE Session on “The Fed’s New Lending Facilities” ASSA Meetings San Francisco

o n . ay or Stanford University  ,

•

•

• •

Summary Excellent   paper   on   economic   rationale   for   PDCF – Analogy:   Broker ‐ Dealers   (PDCF),   Depository   Institutions   (PCF)        – Haircuts,   margins,   leverage,   downward   spirals   (33:1   to   20:1) Levera e   first   rose   in   recent   crisis   SIVs   lar e   rate   cuts Structural   changes   over   time: – Growth   of   broker ‐ dealers   and   hedge   funds   relative   to   epos ory   nst ut ons – Trend   toward   tradable   assets   and   liabilities           Protections:   Penalty   rates,   regulation,   moral   hazard Main   Comments: – Empirical   Impact   – Slippery   slope   to   more   facilities

• •

 

credit  default  swa  rices  of  rimar  dealers  fell   after   the   introduction   of   the   PDCF To   date,   the   PDCF   seems   to   be   having           markets,   as   many   primary   dealers   found   that   their   perceived   risk   of   default,   as   measured   by   credit   default   swap   prices,   declined   after   the   facilitys   introduction.              ,     take   a   look   at   CDS   rates   for   Merrill   Lynch   and   Goldman   Sachs   (for   example)

 

millions of dollars

150,000

125,000

100,000 75,000 50,000

25,000

0

PDCF

CDS Goldman Sachs

basis points 500

Weekly  averages  of  daily  CDS  data  and  weekly  average  PDCF  loans  

400

200

100

0

millions of dollars

150,000 125,000

100,000 75,000 ,

25,000

0

PDCF CDS Merrill Lynch CDS Goldman Sachs

basis points 600

500

400

200

100

0

Daily  CDS  data  with  daily  PDCF  date  interpolated  from  weekly  average  data

•

•

 

 

 

Gra hical  evidence  is  difficult  to  inter ret • CDS   rates   appear   to   increase   when   PDCF   loans   rise! • But   there   are   cause   and   effect   issues   to   sort   out • We   can   apply   time   series   methods   to   sort   out   the   timing   differences ‐ ‐      between   CDS   rates   (on   Merrill   Lynch    and   Goldman   Sachs)   and   PDCF   loans • Compute   impulse   response   functions   from   the   VARs • Look   for   significant   effects     

60

40

20

0

-20

PDCF  and  Goldman  Sachs  CDS  Impulse  Responses Based   on   Weekly   Data Res onse to Choles ne S.D. Innovations ± 2 S.E. Response of CMER1U5 to CMER1U5 Response of CMER1U5 to PDCF 60

40

20

0

-20

-40 -40 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Response of PDCF to CMER1U5 Response of PDCF to PDCF 30,000 30,000

20,000

10,000

0

-10,000

20,000

10,000

0

-10,000

-20,000 -20,000 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

80

0

-40

PDCF  and  Merrill  Lynch  CDS  Impulse  Responses Based   on   Weekly   Data Response to Cholesky One S.D. Innovations ± 2 S.E. Response of CGS1U5 to CGS1U5 Response of CGS1U5 to PDCF 80

0

-40

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Response of PDCF to CGS1U5 Response of PDCF to PDCF 30,000 30,000

20,000

10,000

0

-10,000

20,000

10,000

0

10,000 -

-20,000 -20,000 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

•

•

 

 

  

‐

 

  

 

 

 

 

 

    negatively •     of   the   PDCF CDS   rates   Gran er ‐ cause   PDCF   loans   positively      loans   at   the   time   of   market   stress

 

 

 

 

• • • •

•

•

A  Slippery  Slope  to  More  Facilities?   Securities   (Treasury   and   Agency)   held   outright   Repos   Loans   from   the   TAF Other   Loans Primary   Credit   Facility   (discount   window) – Primary   Dealer   Credit   Facility – Asset   Back   Commercial   Paper   Money   Market   Mutual   Fund   Liquidity   Facility oans   to   – Term   Asset ‐ Backed   Securities   Loan   Facility   (credit   card,   student,   auto)* Private   Portfolio   holdings       – Maiden   Lane   I   (Bear   Stearns) – Maiden   Lane   II    (AIG)      – Money   Market   Investor   Funding   Facility* – Mortgage   Backed   Securities   Purchase   Program*         

 

Note:  Minimal  change  items  are  gold  stock,  SDRs,  Treasury  currency   outstanding,   seasonal   and   secondary   discounts,   float  

 

billions of dollars 900

800

600 500

400

300

100

0

2000

December 31, 2008  $848 Billion

Reserve Balances of De ositor Institutions at Federal Reserve Banks

2002

2004

September 10, 2008  $8 Billion

2006 2008