What life is like as a Scientist in Congress Bill Foster U.S. Congressman (Ret.) Illinois14th...
Transcript of What life is like as a Scientist in Congress Bill Foster U.S. Congressman (Ret.) Illinois14th...
What life is like as a Scientist in Congress
Bill Foster
U.S. Congressman (Ret.)
Illinois14th District
Introduction
Bill Foster
U.S. Congressman (Ret.)
Illinois14th District
OUTLINE1. Bill’s Trajectory:
– Businessman– Physicist– ?? Politician ??
2. Financial Services – Plots & Data
3. Applied Physics: A PID feedback controller for stabilizing Real Estate Housing Bubbles
(Talk at AEI)
Bill Foster’s TimelineI. Businessman: ETC Inc (1974…~1980)
II. Physicist: (1979-2005)– IMB Proton Decay Detector (~1980-1984)– Fermilab (1984-2005)
• HEP Experiments: CDF & SDC• Accelerators
III. Politician (2006 - )– Why would you possibly do such a thing?– Learning to Campaign– A day in the life of a U.S. Congressman
Electronic Theatre Controls, Inc.~1974 -1980
Electronic Theatre Controls, Inc.Today
• 500 Employees in Middleton, WI
IMB Detector 1979 – 1984 (…1987A)• Bill’s PhD Thesis (under Prof. Larry Sulak)• No Proton Decay, but saw SN1987A neutrinos
IMB Proton Decay Experiment
"Did anyone read on the front page of the Times that matter is decaying? Am I the only one who saw that? The universe is gradually breaking down. There's not going to be anything left. I'm not talking about my stupid little films here. Eventually there's not going to be any Beethoven or Shakespeare or ..."
Woody AllenStardust Memories (1980)
Fermilab’sAccelerato
rs
Past,Present,
and Future
Dr. G. William Foster
November 8, 2002
Bill at Fermilab 1984 - 2006
Collider Detector at Fermilab
(CDF)
1984 - 2005
CDF Central Tracking ChamberElectronics High Pt Track Processor
“Damn Fast”
Event Display Program
CDF
High Pt Track Processor
• Hand-wired(Wire-wrapped) FastBus Boards
• Finds all of the High-Pt Tracks before the next crossing
• Essential for Top Quark discovery
e-u top Quark Search & Discovery
QIE Integrated Circuit (invented for Superconducting Super Collider)
• Deadtimeless Charge Integrator and Floating-Point Digitizer
• Still in use today for LHC experiments
Particle Accelerators 1994-2006
FermilabMain Injector
Antiproton Recycler Ring
8 GeV Transfer Line (1996)Fermilab Booster Main Injector
Permanent Magnets
Very Large Hadron Collider (VLHC) 100kA Superconducting Transmission Line Magnet
Beam Damper Circuits for Main Injector
8 GeV Linac / Proton Driver / Project X
The Familial Recessive Gene for Adult-Onset Political Activism
• Bill Got Married,
• Raised a Family,
• Had a reasonably interesting career in Physics.
So: why go into politics?
Bill’s Mother and FatherJeanette Raymond and George William Foster
Met on Capitol Hill ~1950
Bill’s Mom worked for Sen. Paul Douglas (D-IL)
2005 – Bill’s Youngest Child (Christine) Heads off to College
Patrick Murphy 2006 Campaign
Bill Started the 110th Congress as an Intern in the Office of Rep. Patrick
Murphy
• He ended it as a sitting Congressman
Gerrymandering and the IL-14th
IL-14th
Illinois 14th Congressional District
The Campaign Staff
Bill’s Favorite Parade Tie
County Fairs
Churches
FARMERS
Photo Shoot
March 8, 2008
Swearing-In Ceremony – March 11, 2008
2008: An Interesting Year to be a Politician From Illinois
Denver Convention 2008
CongressionalBall - 2008
What Life as a Congressman is Really Like
• Scheduled 12 Hours each day
• Read myself to sleep each night with a ¾” stack of papers…
• Live in a Crummy Efficiency Apartment in DC
• After last votes on Friday, come back home each weekend
(where I get scheduled up by the staff in the District Office…)
Life in Washington
Efficiency Apartment
2 blocks from Capitol
“Congress on Your Corner”
So, why does anyone take this job?
Voting Machine in the U.S. House of Representatives
Supporters
Energy Policy in Congress
Questions About Solar Energy…
Armed Forces
H.R. 3962 - Affordable Health Care for America Act
Financial Services Committee
• Our Economy has Many of the Propertiesof an Oscillating Analog Electronic Circuit
$45T
$50T
$55T
$60T
$65T
$70TJu
n-07
Sep-
07
Dec
-07
Mar
-08
Jun-
08
Sep-
08
Dec
-08
Mar
-09
Jun-
09
Sep-
09
Dec
-09
Mar
-10
Net Worth of U.S. Households
$17.5 TWealth DestroyedOct. 2007 - March 2009
Source: Federal Reserve Flow-of-Funds Data Series Z1 : June, 2010 Rep. Bill Foster http://Foster.house.gov
$6.3 T
RecoveredSince Stimulus
Stimulus SignedFebruary 17, 2009
GreatRecession (Started Q2 2007)
BUSH OBAMA
HOOVER
Great Depression (1929-1933)
FDR
Stimulus
70%
75%
80%
85%
90%
95%
100%
0 0.5 1 1.5 2 2.5 3 3.5 4
Hou
seho
ld N
et W
orth
Com
pare
d to
Pea
k
Years from Peak
Drop in Household Net Worth
Sources: Frederic S. Mishkin, "Household Balance Sheet and The Great Depression," J. Econ.Hist.38 (Dec.1978), p. 920. & Federal Reserve Flow-of-Funds.
Worse than the Great Depression
Actual Intervention
No Intervention
Stimulus Only
Financial Only
$45 T
$47 T
$49 T
$51 T
$53 T
$55 T
$57 T
$59 T
$61 T
$63 T
$65 T
U.S
. Hou
seho
ld N
et W
orth
($T)
Rep. Bill Foster - Aug 2010 Sources: Federal Reserve Board & Moody's Analytics
Household Net Worth Actual vs. No Intervention
$7 TIncrease
due toIntervention
WHAT IF WE HAD DONE NOTHING ?American families would be ~$7T Worse Off
Manufacturing Employment
?1/3 OF JOBS
LOST IN 8 YEARS
STEADY FOR 30 YEARS
10 M
11 M
12 M
13 M
14 M
15 M
16 M
17 M
18 M
19 M
20 M
1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 2013
Man
ufac
turin
g Em
ploy
men
t (m
illio
ns)
Rep. Bill Foster Source: U.S. Bureau of Labor Statistics
U.S. Manufacturing Job Loss - 2001-2009
BANK FAILURES AND SUSPENSIONS 1864-2009
Dodd-FrankWall-Street Reform Bill
• Bill ended up being 10-for-10 in getting his amendments into House-Passed reform bill.
• 6 ½ of these survived the compromise with the Senate.
APPLICATION OF ANALOG ELECTRONICS TO POLITICS
Counter-Cyclical Down Payments:
A Simple Method for Mitigating Future Housing Bubbles
Rep. Bill Foster (IL-14)
AEI - June 12, 2010
With suggestions & comments from: Alan Boyce, Polly Cleveland, Mason Gaffney,
Alex J. Pollock , and Jonathan Reiss.
Counter-Cyclical Down Payments
THE CONCEPT:
• Automatically increase down payments when housing prices start to rise rapidly
Dampen or Eliminate Future Housing Bubbles
No attempt to regulate long term asset values,just damp out boom-bust oscillations.
“AC-coupled feedback loop”
This is a Feedback Loop• Well known from engineering controls theory• Some examples:
– Thermostat– “Governor” on a lawn mower engine (Gibbs)– “Cruise Control” on an automobile
• Regulates fixed speed as car goes up & down hills
– Computerized Accelerator Control on modern cars• programmed to limit acceleration & deceleration for
passenger comfort even while imposing no limit on car’s overall speed
Why Dampen Housing Bubbles?
• That is where the money is:– Housing dominates household net worth– We can survive a stock bubble but not a housing
bubble• Time scale of real estate bubbles is tractable
– Develop over months to years• Low volatility of intrinsic worth of housing
– Not like stock valuations based on volatile earnings– Housing demand changes on demographic time
scales• Good statistical measures exist for feedback
– Large numbers of public transactions in regional housing markets
http://HousingBubble.jparsons.net
A Simple Example Damping Formula
MinimumDown
Payment= +CONSTANT
(e.g.10%)
Percentage Increase inReal Regional Housing
Prices Over Past 3 Years(or a fraction thereof)
NOTES:1. If the housing price increase is negative, it is treated as zero.2. A fixed fraction (e.g.40%) of the price increase could be used.
This controls how strong the damping of bubbles will be.3. There is nothing magical about using the 3 year price rise …
anywhere between 1-5 years works OK.4. LTV on Home Equity Loans, etc. would be similarly limited.5. The CONSTANT term is subject to socio-political debate and
will not be discussed here. Here we assume 10%.
Can be thought of in several ways:
1. As a feedback mechanism.
2. As a prudent countercyclical principle:“All mortgages have to make sense even if housing
prices retreat to where they were a few years ago”
3. As a simple underwriting rule:“I will give you a mortgage against 90% of the long-
term value of the property, but against only a small fraction of the recently-appreciated value.”
Treats skeptically the value of recently-appreciated assets!
How This Would Have Worked
• The following plots show how this formula would have applied to price bubbles in regional markets over the last 20 years.
• In reality, the big increases in counter-cyclical down payments would have absorbed speculative capital and largely squelched the bubbles.
• There are also higher-order psychological effects since even the specter of large counter-cyclical down payments would have discouraged any large inrush of speculative capital.
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
Las Vegas, NV (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=55%
Down Payment (3yr. Diff.) (Fract=50%) Peak=32%
10% Fixed Down Payment
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
Seattle, WA (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=38%
Down Payment (3yr. Diff.) (Fract=50%) Peak=24%
10% Fixed Down Payment
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
Cleveland, OH (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=17%
Down Payment (3yr. Diff.) (Fract=50%) Peak=14%
10% Fixed Down Payment
ORDWAS on Air Force One
BACKUP: DETAILS ON COUNTER-CYCLICAL LOAN-
TO-VALUE LIMITS
• OPTIMIZATION OF FEEDBACK GAIN
• BEHAVIOR IN REGIONAL MARKETS
• LEGISLATIVE IMPLEMENTATION
Input Data
• Housing Price Data (S&P / Case-Shiller)– Seasonally Adjusted to avoid yearly ripples– Regional since markets have individual
bubbles
• Inflation Adjustment using CPI-U– excluding Food and Energy (less noisy)
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
Miami, FL (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=53%
Down Payment (3yr. Diff.) (Fract=50%) Peak=32%
10% Fixed Down Payment
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
Minneapolis, MN (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=34%
Down Payment (3yr. Diff.) (Fract=50%) Peak=22%
10% Fixed Down Payment
0
50
100
150
200
250
300
Infla
tion-
Adju
sted
Reg
iona
l Hou
sing
Pric
es (2
000 =
100)
Countercyclical Down Payment Based onRecent Rise in Regional Real Estate Price
10-City Average (Inflation Adj.)
Down Payment (3 yr. Diff.) (Fract=100%) Peak=43%
Down Payment (3yr. Diff.) (Fract=50%) Peak=27%
10% Fixed Down Payment
Countercyclical Down Payment Absorbs Capital from House
Flipping
Normal Flipping
Flipping with Countercyclical Down Payment
Initial Equity $10,000 $10,000 Down Payment 10% 10% Affordable House Price $100,000 $100,000 Market Appreciation 25% 25% Appreciated House Value $125,000 $125,000 Equity after Sale $35,000 $35,000 Down Payment for New House 10% 35% See Note Affordable New House $350,000 $100,000
Note: Counter-Cyclical Down Payment = 10% + 25% Market Appreciation
More Optimizations of the Feedback Loop
• What variables are best for feedback sensing:– Housing prices: (national, regional, sectoral?)– Housing volume: (sales volume or construction?)– Inflation-adjusted? Subtract long term trends?– Price/Rent ratios?– Incomes & creditworthiness of buyers?– Geographical smoothing?
• What actuators are we using for feed back:– Loan-to-Value (LTV) limits– Other underwriting requirements– Securitization “skin-in-the-game” for MBS
Optimizing the Gain of the Feedback Loop
(Ideas from Jonathan Reiss)Q: What is the optimum fraction of the price rise
to apply to the down payment (i.e. the gain of the feedback loop)?– Too small, and it will be ineffective– Too large, and it may over-correct and oscillate– Jonathan Reiss estimates ~ 40%– Needs modeling… is there calibration data
anywhere? Fortunately, damping feedback loops are
very insensitive to the exact value of the feedback gain (at least in electrical engineering).
Formula vs. Omniscient Regulator
• Similar feedback could be implemented via an omniscient and independent real estate market regulator (presumably the Fed).– The Chinese & Israelis are doing this today.
study this as source of calibration data?
• Our approach of using a formula is:– More Transparent and Predictable
– Simpler
– Less prone to political interference
Life Inside the Feedback Loop
• When real estate markets slow down, The Lending Tree™ will start running advertisements talking about how NOW is the right time to buy a house because housing prices are not rising as fast as normal and down payments are the lowest in years!
These ads will disappear as soon as a bubble starts.
Squelching the Bubble Psychology
• Negative feedback removes the most powerful element in the bubble psychology:
Nobody likes to be the dumb guy in the room who missed out on the big price run-up.
• People will know that even if there is a big price run-up, counter-cyclical LTV’s will remove potential speculative buyers from the market and you won’t be able to flip it.
Legislative Implementation
• As an underwriting standard:– Fed authority to set nationwide mortgage
origination standards (for consumer protection) broadened to include LTV limits for market protection
• As a condition on government-backed MBS– Currently dominate the market
• but Fannie & Freddie will have to be unwound.
– Do not have to control 100% of the mortgages issued to dampen a housing bubble
…but you still have to compete.
Some Down Sides of Regulating Bubbles
1. Will slow down the re-deployment of capital in response to real changes in the economy– The misallocation of capital due to this effect is
tiny compared to the recent housing bubble.
2. Will distort the economy by punishing one sector for a “nearby” bubble– “Gary is not Chicago”– Will be small under the non-bubble conditions
that should be typical with bubble damping
3. Politicization of the housing indexes