The 6am cut - Alphaville by email

Most Popular Posts

  1. The Weekender
  2. US covered bonds? Frederick II would be turning in his grave
  3. Fannie and Freddie link fest - updated with key US Treasury pdfs
  4. Further Fannie and Freddie
  5. Dr Doom: More gloom, the Kondratieff wave and what comes next
  6. Show more...
  7. Show less...
  8.  

Blogs we're reading

Classified Jobs

Director of Finance and Corporate Services
Recruiter: NSPCC
Head of Customer Service - Savings
Recruiter: Nationwide Building Society
Business Analyst
Recruiter: Credit Suisse
Head of Finance
Recruiter: ESCP-EAP
Group Chief Accountant
Recruiter: Misys
Lead Economist, Development Research Department
Recruiter: African Development Bank
Finance Manager
Recruiter: London 2012
Group Manufacturing Leadership Programme
Recruiter: The Royal Bank of Scotland Group

Site Navigation


Principal content

“A trillion dollar mean reversion”

In part, the current financial crisis - or at least, the hitherto dominant, structured finance part of it - was caused by an unhealthy trust in concepts like mean reversion.

That is, the assumption that over time, fluctuations in price “revert” to a long-term mean average.
Mean reversion was used as a key metric in the modelling of many structured instruments’ default risk or pricing. Too much reliance on mean reversion underplayed the likelihood of a sustained, highly correlated financial crisis.

Ironic then, this, from a note sent out today by Deutsche bank’s Jim Reid:

Banks' cumulative losses

The chart shows “excess” profits earned by US financials relative to their long-term relationship to US GDP. That is, profits which deviate from the mean projection. In other words, banks profits have not mean reverted - until now.

Here’s another graph, which rather clearly illustrates the cleavage between banks’ profit trajectories and that of regular corporates:

Jim Reid banks profit

The irony, then, being, that what we are witnessing in the current financial sector could be described - in financial modelling terms - as a rather dramatic mean reversion. In modelling CDOs banks mean reverted too much. In their own performance, they mean reverted too little.

The US Financial sector has made around 1.2 Trillion ($1,200bn) of “excess” profits in the last decade relative to nominal GDP.

So mean reversion would suggest that $1.2 trillion of profits need to be wiped out before the US financial sector can be cleansed of the excesses of the last decade.

Clearly this is too extreme if you believe that the financial sector has seen a sustainable structural change in its business model over the last decade or that it is uniquely positioned to exploit strong Global growth. Sectors do change/grow in importance over time… However, this is the kind of argument that has helped lead to this credit crisis. The financial sector has grown rapidly over the past decade, has embraced high leverage, and we’ve also seen a shadow banking system emerge that didn’t really exist a decade or so ago. Calculating the “natural” appropriate size for the financial sector relative to the rest of the economy is a phenomenally difficult conundrum.

A conundrum we will leave others to solve.

This entry was posted by Sam Jones on Tuesday, July 15th, 2008 at 12:09 and is filed under Capital markets, Hedge funds. You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.

Email this story
Add to Del.icio.us
Share on Facebook
Digg this post
RSS Feed

Comments

  1. Jul 17   23:11 Posted by KevX [report]

    Washout is needed and inevitable. Application of merchant banking ethos in
    inappropriate situations and, oh yeah, a little self-dealing.

    Don’t worry next big liquidity wave is coming with Boomer’s retirement money however, cannot be managed in same way by IBs and Wall St.. Will take real fiduciary business practices. Yikes!! How do I trade that?

  2. Jul 15   18:05 Posted by fugger [report]

    Fourth theory is probably the most significant- banks are just highly leveraged bets on economic growth ie high beta. A key element of the 2000s was that they became ever more leveraged as they learnt to arbitrage capital adequacy rules eg sticking everything in the “trading book”

  3. Jul 15   17:35 Posted by anonymous [report]

    The graph shows illusionary “Financial Profits”, which have historically been made up numbers based on DIY valuation models. It does, of course, not change the fact that at some point the made up numbers need to be replaced by the real ones. Mainly due to the fact that the underlying assets need to provide real cash flows over time. Or do they? Can we keep living in a dream world?

  4. Jul 15   15:35 Posted by pegnu [report]

    and frankly they’re lucky most of the population will never understand the magnitude of the robbery.

  5. Jul 15   15:19 Posted by pegnu [report]

    the financial services sector is clearly going to be shrinking quite dramatically.

  6. Jul 15   14:18 Posted by fugger [report]

    Why were banks making excess profits? Three theories:
    - booking profits on illusory asset price gains and unsustainable loans where losses come later
    - barriers to entry
    - implicit/explicit government guarantee and ineffective regulation to ‘correct’ for that
    The first one is reversing but the third one is probably getting stronger - and ironically securitisation probably reduced barriers to entry, which are now rising again. So look out for the strong bank survivors with less competition and a cast iron government guarantee

  7. Jul 15   13:42 Posted by wdm, ny [report]

    Much of the “profit” since 2001 has just been markup of assets still on the books which are in the process of getting marked back down…very early stages.

  8. Jul 15   13:32 Posted by burnt quant [report]

    @ G Cox

    You are quite right to call me to task with my inappropriate smiley.

    Just nice to know that I’m not the only one taking hits in the market sometimes, and that the big boys aren’t escaping either.

  9. Jul 15   13:18 Posted by VP [report]

    I’m with GCox although…

    If house prices can do 20% ABOVE the rise in average earnings in a year, then no reason they can’t overshoot as far in the other direction.. (other than that they have a utility value not possessed by eg equities).

  10. Jul 15   12:42 Posted by G Cox [report]

    Anonymous, if you are talking of mean reversion of house prices then you must think of reversion to 98 in today’s prices or better still nominal incomes otherwise it makes no sense.

    The profit numbers above are reverted to profit inflated by nominal GDP for example.

  11. Jul 15   12:33 Posted by G Cox [report]

    Burnt Quant. What about ‘credit reversion’.
    Sounds more elegant than de-leveraging.

    ( PS No smiley faces please. Some poor people are losing large chunks of their retirement income.)

  12. Jul 15   12:14 Posted by Anonymous [report]

    Oh, well, I still believe in mean reversion. I am confident that UK house prices will revert to 1998 levels (the start of the boom) in 1-2 years. Bubbles always pop a lot faster than they inflate.

  13. Jul 15   12:13 Posted by burnt quant [report]

    We’re certainly seeing some “mean-reversion” in the cost of credit aren’t we?

    In fact I refuse to call this a credit crunch, instead I shall henceforth refer to it as the “credit normalisation”.

    Given that estimates for the hit to the (global) banking sector started off at $100bio and the last (pessimistic) figure I saw quoted was $1.6trio we might yet come somewhere close to eroding those $1.2trio of excess profits ;-)

This post is closed to further comments.