Monday, December 08, 2008

Occam’s Razor and the Economy

When the Numbers Don’t Add Up...

...check who’s doing the adding.


That should be the motto for all Americans as politicians, business executives and economists pitch their solutions for reversing (or at least halting) the current economic collapse. Virtually all players involved in the system now all claim that the overwhelming complexity of our "modern" financial system and the blizzard of new-fangled financial instruments made anticipating the timing and magnitude of any potential correction within the system almost impossible. Given the magnitude of the problem that’s now evident, the unpredictability seems highly suspect in retrospect.

Occam’s Razor is a term for an approach to solving problems which attempts to avoid distractions by eliminating factors that have no observable or proven impact on the phenomena being explained or problem being solved. In plain language, it basically says when faced with a choice between an explanation or theory that could be a cross between a Rube Goldberg machine (#1) and M.C. Escher drawing (#2) versus an explanation involving five variables and basic arithmetic, the simple solution with the arithmetic might be the better starting point.

It turns out Rube Goldberg machines, Escher drawings and Occam’s Razor – more specifically, the failure to apply the principle to information provided us – have everything to do with the meltdown we’re now experiencing. Experts and individual consumers alike both saw signs that segments of the economy weren’t performing well yet classic measures of those segments weren’t "adding up":

* consumers saw low reported inflation, yet skyrocketing housing prices
* consumers were granted ever larger lines of credit despite a lack of growth in real income
* consumers enjoyed the trappings of higher levels of wealth but their credit balances increased year after year
* experts saw vast bundles of securities of unverified provenance and divergent risk levels blended together and magically emerge with top quality AAA ratings
* experts saw massive credit injections towards the end of the cycle, yet saw little increase in reported IN-flation
* experts now see massive contractions in lending, yet don’t see the expected DE-flation

It’s certainly possible that the explanation for all of these ordinarily contradictory signs of economic health boils down to an otherwise PERFECTLY healthy economy with an unexpectedly high level of positive feedback between the second order derivative of increased lending (or the rate of change in the rate of change of lending) and its impact on future credit evaluations which sparked a change in risk assessment which altered behavior of consumers who decided to defer new car purchases for 15.7 months, shifting the demand curve out in time on the manufacturing sector which triggered a faster than expected loss of jobs which triggered mortgage defaults which triggered the collapse of the derivate markets which … ZZZZZZZZZZZZZZZZZZ

Or…

It’s possible that virtually all of the damage arose from a single (intentional) change in procedures used to fudge --- I mean “report” --- inflation back in the 1980s. It’s possible that the economy has been collapsing for over two years but recognition of the effort in the ER on the part of the Federal Reserve to revive the patient was delayed by a single (intentional) change in statistical reporting in March 2006.

Fudging Inflation

In the case of inflation reporting, a decision made during the Reagan Administration in 1983 alter inflation calculations by replacing actual dollars spent on home mortgages with a number referred to as “owner’s equivalent rent”. Documentation of the original public or actual justification for this change is hard to find. Maybe they thought rapidly rising home prices acted as “investments” for homeowners (like stock appreciation) and therefore only the “real” value of the home – whatever THAT is – should be counted. More cynically, maybe this was a ploy to create a quick win against “inflation”, reduce union expectations of future inflation, and reduce growth in labor costs. Maybe it was a ploy to begin cutting growth in payments for CPI based entitlements, a ploy formally adopted by Clinton and Greenspan in the 1990s.

The motivation for the change might be fuzzy but the impact on inflation statistics every since could not be more clear. (#3)

* the housing portion of inflation has never equated to more than 6% since 1983
* prior to the change, housing prices contributed nearly HALF of the total rate of inflation
* by 2008, housing prices only contributed roughly 25% of the drastically massaged, drastically lower inflation reported (see the graph at the link)

Fudging the Money Supply

In the taxonomy of money, M0 is physical currency sitting in bank vaults and individual wallets. M1 includes M0 plus balances of checking accounts and travelers checks. M2 includes M1 plus savings account balances, CDs less than $100,000 and money market account balances. M3 is M2 plus CDs over $100,000 and other institutional accounts and loans between banks. When the overall economy experiences real (non-inflationary) growth, the M3 number which reflects all “money” in the system will grow roughly in proportion to the real growth rate of the economy.

Those following money supply statistics were thrown a curve ball on March 23, 2006 when the Federal Reserve Bank decided to stop reporting M3 statistics. The rationale cited by the Fed at the time now seems criminally negligent or stupid in hindsight (#4):

M3 does not appear to convey any additional information about economic activity that is not already embodied in M2 and has not played a role in the monetary policy process for many years. Consequently, the Board judged that the costs of collecting the underlying data and publishing M3 outweigh the benefits.

Hmmmm. ”has not played a role in the monetary policy process for many years.” Maybe not but the decision to STOP reporting it precisely in March of 2006 appears to be a definite part of the Fed’s strategy in attempting to prop up a collapsing system without spooking the herd about skyrocketing amounts of dollars entering the money supply through massive credit infusions to member banks.

What did the change mean for investors and consumers? Using consistent numbers for the money supply would have reflected more “dollars” in the economy chasing the same goods, which is the definition of inflation. If you thought your 3 percent raise kept you at bay with inflation, you were actually losing ground. If you thought investment gains of 7-10 percent still put you three or four percent ahead in real terms, you actually may have lost money. Instead of interest rates jumping upward sharply and providing a signal to marginal borrowers to avoid taking on additional debt, interest rates remained relatively low, tempting more borrowers into the pool.


(Un-) Do the Math

Neither of these changes is being discussed here because they are recent news. Both were widely reported when they took place. Still, no one lobbied to undo them, even when personal or professional experience (housing prices doubling but 5% inflation??? Triple-A rated investments with skyrocketing CDO risk premiums???) clearly indicated something was wrong.

Why?

Maybe Americans were just stupid enough to not realize this WASN’T rocket science and that one didn’t need calculus and a degree in Finance to realize housing can’t double in five years and only see 3 percent inflation. Maybe Greenspan successfully hypnotized politicians and bankers with Rube Goldberg language and bored everyone to the point of never going back to basic arithmetic to figure out we’re going broke one HELOC and sub-prime loan at a time.

The lesson should be painfully clear now. No action taken by a major corporation or Congress requires analysis beyond a five year time horizon or math beyond what you can do on a simple calculator to determine if it makes sense. No one in a position to make most of these decisions sees more than a 10-page PowerPoint presentation justifying a decision and even that ten pages includes a title page and table of contents. If you can’t understand the revenues and expenses or costs and benefits of a solution after a ten minute explanation, it’s probably not you, it’s the model or the person proposing it that needs work.

If Congress really wanted to lay the groundwork for improving trust in the markets and liquidity in the financial system, there is no better place to start than undoing all of the bogus adjustments and outright distortions made to the numbers that act as the dashboard for the economy. There may be no more important numbers to correct than the data for inflation and the money supply. Until the distortion of these numbers is corrected, it’s impossible to calculate the true “value” of any long term proposal for fixing the economy.


=============================

#1) http://www.rubegoldberg.com/gallery_02.php

#2) http://www.mcescher.com/Gallery/back-bmp/LW389.jpg

#3) http://www.marketoracle.co.uk/Article4018.html

#4) http://www.federalreserve.gov/releases/h6/discm3.htm