Upload
others
View
10
Download
0
Embed Size (px)
Citation preview
Presentation to the U.S. Treasury Presentation to the U.S. Treasury and the Treasury Borrowing and the Treasury Borrowing
Advisory CommitteeAdvisory Committee
May 3, 2005
QuestionQuestion
Please discuss the characteristics of Treasury liability portfolio including average maturity of debt, steady state issuance and rollover. Do these metrics adequately capture the Treasury’s policy concerns? Are there other metrics that we should be using to develop debt management policy?
What Does Treasury Liability What Does Treasury Liability Look Like?Look Like?
Source: Treasury Chart Book, Merrill Lynch
Debt PortfolioAverage maturity of issuance has stabilized at roughly 3 yearsMaturity of total outstanding drops from 4.6 years to less than 4 years over the next 5 yearsComposition of nominal issuance is within historical rangesCurrent issuance patterns lead to a growing proportion of 5-year notes and TIPS in the portfolioThe percent of debt maturing with 3 years or less to maturity is projected to remain stable at slightly more than 60%
Distribution of Marketable Debt Outstanding
0%
5%10%
15%
20%
25%30%
35%
1980 1987 1994 2001 20080%
5%10%
15%
20%
25%30%
35%
Bills 4-7 yrs 10 yrsBonds TIPS 2-3 yrs
Pro
ject
ed
Debt Maturity Measures
0
20
40
60
80
100
80 84 88 92 96 00 04 08
Mon
ths
0
20
40
60
80
100
Average maturity of issuanceAverage maturity
Pro
ject
ed
Average Maturity of Debt
0
24
6
8
1012
14
85 87 89 91 93 95 97 99 01 03
Year
s
0
24
6
8
1012
14
U.S. CanadaFrance GermanyU.K. JapanItaly
Steady StateSteady State
Source: Treasury Chart Book
Bills, acting as residual borrowing device, start to increase in 2008. Need to decide if more permanent, longer-term issuance is warranted
Treasury AnnualNet Market Borrowing
-300
-200
-100
0
100
200
300
400
500
98 00 02 04 06 08
$ Bi
llion
s
Bills $ billions 2-under 5 years5-10 years Over 10 yearsBuybacks
Projected
Financing Residuals GivenCurrent Issuance Pattern
-300
-200
-100
0
100
200
300
400
500
600
700
2005 2006 2007 2008 2009
$ Bi
llion
s
-300
-200
-100
0
100
200
300
400
500
600
700
Plus average absolute error in OMB estimateOMB MSR FY2005 Budget--- Net FinancingMinus average absolute error in OMB estimate
Bars indicate estimated change in f inancing required in given year for different deficit outcomes
Rollover RiskRollover Risk
Source: Treasury Chart Book
Rollover risk not high by historic standards
Ill-defined, not comparable to HY / EMG world
Percentage of Debt Maturing in Next 12 to 36 Months
25%
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
80 82 84 86 88 90 92 94 96 98 00 02 04 06 0825%
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
Maturing in 12 months Maturing in 24 months Maturing in 36 months
Objective DefinedObjective Defined
Primary goal: Finance the government borrowing needs at the lowest cost over time
Issue debt in a regular and predictable manner, provide transparency…, seek continuous improvements in the auction process
Fisher “Treasury does not try to outsmart the market at any one moment, or (try) to be a “market timer” with respect to any particular shape of the yield curve”
Commentary, How Well Do They Commentary, How Well Do They Capture Policy ConcernsCapture Policy Concerns
On balance, Treasury portfolio appears to be well balanced. Designed to meet the Treasuries objectives and provide flexibility for most possible outcomes
Assymetric risk profile, higher than expected borrowing needs will force Treasury action well before lower needs
Source: Goldman Sachs, Lehman Brothers
Federal Government Liabilities
-600
-400
-200
0
200
400
600
800
90 92 94 96 98 00 02 04
$ Bi
llion
s
2500
3000
3500
4000
4500
5000
5500
$ Billions
Net borrowing (Flow) LHSLiabilities (Stock) RHS
Budget Balance
-500
-400
-300
-200
-100
0
100
200
300
80 83 86 89 92 95 98 01 04
$ Bi
llion
s
20yr 4yr 7yr
3yr 1yr
30yrTIPS
4wk
Buybacks 3yr
5yr, 20yr TIPS
Demand and Traditional Auction Demand and Traditional Auction StatisticsStatistics
Source: Wrightson
Traditional auction statistics (bid / cover ratio and tail) don’t help capture the demand functionOnly very low correlations with auction sizesDemand function unfortunately does not lend itself to precise calculations
TIPs Auctions
0
1
2
3
4
5
6
0 5 10 15Gross Issue ($bn)
Cov
er R
atio
2yr Notes
0
0.5
1
1.5
2
2.5
3
3.5
0 10 20 30Gross Issue ($bn)
Cov
er R
atio
3yr Notes
0
0.5
1
1.52
2.5
3
3.5
0 10 20 30Gross Issue ($bn)
Cov
er R
atio
5yr Notes
0
0.51
1.5
2
2.53
3.5
0 10 20 30Gross Issue ($bn)
Cov
er R
atio
Demand and Foreign Demand and Foreign ParticipationParticipation
Source: Treasury Chart Book, Goldman Sachs1/ Privately held debt excludes holdings of the Federal Reserve2/ Series for estimated foreign holdings. See www.treas.gov/tic/index.html for source data3/ Source: Federal Reserve Bank of New York statistical release H4.1* Note:April 05 reading is the month to April 15
Foreign participation in Primary and Secondary markets can be unpredictableRecommend better data collection, focus on long-term changes in their lending preferences
Foreign Holdings as a Percentof Total Privately Held Public Debt 1/
53%
29%
0%
10%
20%
30%
40%
50%
60%
93 95 97 99 01 03 050%
10%
20%
30%
40%
50%
60%
Estimated foreign holdings as % ofprivate. 2/Foreign and International InstitutionalHoldings at FRBNY as % of private. 3/
BenchmarkRevisions
Central Bank Purchases Drop in April*
-10
0
10
20
30
40
Jan-03 Jul-03 Jan-04 Jul-04 Jan-05
US$
Bn
-10
0
10
20
30
40
TreasuriesTreasuries+Agencies
Montlhy Change in US Security Holdings by Official Institutions
TICsData
Fed Data
Less Treasury Buying in 2005, More Agencies
0
50
100
150
200
250
300
Treasuries Agencies Total
Chg
in O
ffici
al H
oldi
ngs
US$
Bill
ion
2004 Q1 2004 Q1 2005
Demand and the Rest of the Demand and the Rest of the Fixed Income UniverseFixed Income Universe
Source: Citigroup, Lehman Brothers, PIMCO
Demand for Treasuries can be effected by the alternative Fixed Income Universe SpaceExample: A secular shift in the mortgage market could increase the demand for Treasuries relative to all other assets if demandfor duration stays constant
ARMs as Percentage of Overall MBS Universe
Citigroup BIG Index vs. Citigroup BIG ex-Treasury Index
3
3.5
4
4.5
5
5.5
89 91 93 95 97 99 01 03 053
3.5
4
4.5
5
5.5
BIG Duration ex-TreasuryBIG Duration
Effective Duration vs. Effective Duration ex-Treasury
3.0
3.5
4.0
4.5
5.0
5.5
98 99 00 01 02 03 043.0
3.5
4.0
4.5
5.0
5.5
Aggregate Effective DurationAggregate Effective Duration ex-Treasury
1994
1%
12%
2%
85%
Fixed Rate 1/1 Arm3/1 Arm 5/1 Arm
2005
76%
5%
10%
3%
6%
Fixed Rate 1/1 Arm3/1 Arm 5/1 Arm7/1 Arm
Demand and Asset AllocationDemand and Asset Allocation
Source: Federal Reserve Flow of Funds Report - Table L.119.b, The Economics of Private Pensions
Shifts is asset allocation can can have a dramatic impact on the demand for TreasuriesCurrent pension fund dynamics strongly arguing for better asset-liability management, perhaps increasing demand for long-term securities
Asset Allocation
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
46 54 62 70 78 86 94 02Cash Bonds Equity Other
Asset Allocation
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
46 54 62 70 78 86 94 02Bonds Other
Demand and the Yield CurveDemand and the Yield Curve
Source: Bloomberg Financial Markets, RBS Greenwich
Yield curve should give Treasury a sense of the changing maturity preferencesSimple regression of yield curve implies demand for longer maturities has increased, but will it persist?Liquidity for 30yr futures still far below intermediate maturities
Tsy 5-30s Curve Slope Normalised vs. 2y Level Monthly Data 3/1/77-4/1/05
-1.5
-1.0
-0.5
0.0
0.5
1.0
Jan-77 Sep-90 May-04
Perc
ent
y = -0.18123 * x + 1.827791
Curve too flat
CBOT TY Open Interest vs.CBOT Total Open Interest
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
1986 1990 1994 1998 2002
CBOT TY total open interestCBOT bond total open interest
Reality Check and the Reality Check and the 30 Year Bond30 Year Bond
Source: Lehman Brothers
Focus on 30yr Bond issuance disproportionate to potential impact
Market Value % Market Value % Duration
LBAG 8,188,057.00$ 100.0% 100%
Treasury 2,078,841.00$ 25.4% 30.4%
Long Treasury 516,270.00$ 6.3% 15.5%
Annual 30 Year Issuance 30,000.00$ 0.4% 1.3%
Presentation to the Treasury Presentation to the Treasury Borrowing Advisory CommitteeBorrowing Advisory Committee
May, 2005
Question:Question:
Please describe any trends in the Treasury market that you believe are significant to the Treasury as an issuer.
Long Duration Long Duration Supply/Demand ImbalanceSupply/Demand Imbalance
Pension Fund Reform: Well Understood Basics
Long dated assets as a % of Total Pension Fund Assets
0%5%
10%15%
20%25%
30%35%
40%
US UK
At the end of 2004 the median asset allocation of Defined Benefit Plans was 60% equities and only 28% bonds.The PBGC indicated that pension plans are under funded by $450b, a sharp contrast to an over funded position as recently as 2000.Pension plans have large duration mismatches between liabilitieswhich average 12-16 years and assets which average approximately five years. Addressing this mismatch will create demand for long duration fixed income instruments, the availability of which is low relative to other countries such as the UK. Theoretically, the extra demand for dollar duration from the reform proposal could be as high as $2 trillion assuming a 5-year duration mismatch, and over $6 trillion assuming a 10-year duration mismatch. Key point is that pension reform could, to a greater or lesser degree, create demand for longer duration assets. Under-funded liabilities in pension funds is likely to create long duration demand even in the absence of reform given the implicitguarantee provided by the PBGC, backed by the treasurySupply/Demand imbalance further compounds curve flattening, creating a possible curve inversion between 10-year and 30-year maturities.
Potential Rise in Duration Demand ($ Trillion)
$-
$1.0
$2.0
$3.0
$4.0
$5.0
$6.0
$7.0
5yr 10yr
Demand for long-dated cash 10s/30s Spread-of-Spreads
Since the Department of Labor released its proposal for pension reform on January 10, 2005, the demand for long-dated cash securities outpaced the demand for intermediate-dated cash instruments, and outpaced the demand for longer duration derivative assets. Undoubtedly, the majority of this buying originated from speculative players anticipating a future pension bid.Regardless, while swaps can be traded limitlessly, this does not suggest they serve as a perfect substitute for cash securities.
10s/30s spread-of-spreads (in basis points)
(30)(25)
(20)(15)(10)
(5)-
510
8/10/0
0
12/10/0
0
4/10/0
1
8/10/0
1
12/10/0
1
4/10/0
2
8/10/0
2
12/10
/02
4/10/0
3
8/10/0
3
12/10
/03
4/10/0
4
8/10/0
4
12/10
/04
4/10/0
5
Outstanding Mortgage Durations Already Short are Shortening Further.
Real Fed Funds vs. 2s/10s Slope-2.00
-1.00
0.00
1.00
2.00
3.00
4.00
5.00
6.001990 1992 1994 1996 1998 2000 2002 2004
-150
-50
50
150
250
350
450
Real Fed Funds(inverted axis)2s/30s Slope
Maturity Runoff Model of Conforming Balance Mortgage Originations
0.0
5.0
10.0
15.0
20.0
25.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 1516
-30
Years
Tota
l %
Adjustable rate mortgages share of new mortgage origination hassurged from 20% to 60% over the past 2 years. High home prices and rising interest rates could keep demand for ARM products high relative to the total portfolio mix. A maturity runoff model of conforming balance mortgage originations shows the asset liability match frontloaded. Net shortening of mortgage assets effectively reduce the total supply of longer duration debt. Increased supply at the short end of the curve is yet another contributing factor to the flattening trend that has developed in the current tightening cycle.
Long-Term Agency Issuance and Outstanding Treasury Duration
500
600
700
800
900
1,000
1,100
1,200
2001 2002 2003 2004 2005(estimated)
Agen
cy Is
suan
ce -
$ bi
llion
s
55575961636567697173
Tsy
Dura
tion
- Mon
ths
Agency IssuanceTsy Duration
Pool of Longer Dated Instruments Diminishing in Treasury and Agency Market
Fannie/Freddie Mortgage Holdings
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
0
200
400
600
800
1000
1200
1400% of MBS Market
$ MBS (in billions)
Average maturity of outstanding treasury debt has steadily diminished over the last several years. Assuming current issuance patterns of an average 3-year maturity are maintained, maturity of outstanding treasury debt could diminish to four years by the end of 2010. Over the past several years long term agency issuance filled a need for longer dated instruments. Average issuance of long term debt has been approximately 4 1/2 years. Trend decline in GSE balance sheets lessens supply of that closely comparable debt, creating less low-risk, non-call debt in the world; more callable and somewhat messy MBS to absorb along with their estimation risk. If you accept the notion that the market thought of GSE debt as essentially “governments”, then one could envision a widening of swap and corporate spreads as fewer bond portfolios find MBS as suitable investments (either because the bonds are callable or because they are unwilling to accept the associated estimation risk) and choose to reach for the lowest-risk debt class.
Why Not Longer Duration TIPS?
Despite the expansion of TIPS issuance and the introduction last year of both 5-year and 20-year issues, the market remains illiquid relative to nominal bonds.The lack of liquidity is a result of several factors: Low primary dealership participation; a lack of hedging instruments that provide accurate protection for risk; thin secondary trading in the inter-dealer market and high investor concentration.The TIPS market needs further maturation before it can be a viable alternative for nominal securities. In addition, the concentration of ownership among a small group of money managers implies any meaningful portfolio adjustment could lead to market dislocations.
TIPS Daily Trading Volume Relative to Total Treasury Daily Trading Volume
0.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
2002
2002
2002
2002
2003
2003
2003
2003
2004
2004
2004
2004
2005
The Curve and Financing Costs
A simplified look at reducing financing costs by reducing 5-year issuance, and substituting with a mix of short and long-term issues illustrates that since 2001, the treasury would have benefited from lower funding costs by including 30-yr bonds in the mix of assets issued for funding. The chart uses a duration weighted combination of 2-yr and 30-yr bonds to replace reduced 5-year issuance ($100mm 5s = $20mm Bonds + $80mm 2s).Negative numbers show the extra cost, in basis points, incurred by issuing 5-year notes instead of a combination of 2-year notes and 30-year bonds.
2s/5s/30s Butterfly
-100
-80
-60
-40
-20
0
20
40
1/30/9
9
7/30/9
9
1/30/0
0
7/30/0
0
1/30/0
1
7/30/0
1
1/30/0
2
7/30/0
2
1/30/0
3
7/30/0
3
1/30/0
4
7/30/0
4
1/30/0
5
yiel
d of
win
gs -
yiel
d of
bod
y
Rolling Average of 2s/5s/30s Butterfly
Using a one year rolling average of the 2s/5s/30s butterfly, theTreasury’s current savings would amount to roughly 37 basis points.Predicting when a change in issuing patterns is necessary to maintain low borrowing costs is a difficult challenge. The decision to eliminate the 30-year bond in October 2001, made sense in light of budget surplus estimates, as did the elimination of the 4-year, 7-year and 20-year given the expensive funding levels incurred relative to other maturities. The 5-year and 10-year sectors of the curve are expensive funding points compared to various mixes that include a 30-year, and current supply/demand dynamics argue that demand for long duration assets far exceeds the supply and the possible savings for a reintroduction of the bond would be sustainable.
1 Year Rolling Average Savings (bps)
01020304050607080
4/22/2
004
5/22/2
004
6/22/2
004
7/22/2
004
8/22/2
004
9/22/2
004
10/22
/2004
11/22
/2004
12/22
/2004
1/22/2
005
2/22/2
005
3/22/2
005
Foreign Purchases of U.S. Foreign Purchases of U.S. TreasuriesTreasuries
Asian central bank purchases represent 75% of total foreign official purchases. China is the second largest holder of Treasury securities.Revaluation of China’s currency (RMB) would reduce USD weighting in portfolio from 80% toward 65%, which would imply a continued deceleration in purchases of USD assets, including treasuries. Net purchases from Asian accounts over past twelve months have decelerated to approximately $140b, half the sum in the previous 12 month period. Revaluation could have knock–on effects on other Asian portfolios that also manage currency to the Yuan peg.The treasury market could absorb a deceleration in both purchases from China (approximately $50b a year), and Asian buyers more broadly, should there be a shift in exchange rate policy, albeit at higher rates.Reduced buying would exert upside pressure on rates, which could be magnified by a weaker USD, at least over the short term, raising borrowing costs for the Treasury.
Chinese Buying: Steady, but Facing Some New Questions.
Monthly Change in Foreign Holdings in $ billions (6m Moving Average)
-10.0
-5.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
2002 2003 2004 2005
ChinaJapanUK
Weak buying from Japan reflects portfolio reallocation into Agency securities. Share of Agency market has risen from 10% to 18% over the past year.Foreign purchases have more than kept pace with the incremental borrowing needs in the post surplus Treasury environment. Net purchases have averaged almost $30b a month over the past year.Foreign central banks still account for half of those purchases despite outright selling from Japan over the last six months.
Foreign Purchases of US Treasuries
-0.40-0.200.000.200.400.600.801.001.20
1995
19961997
19981999
20002001
20022003
2004
% of total PortfolioInvestmentmean since 1995
Source: EcoWin
Japanese Purchases Declining, but Overall Foreign Buying Remains Robust.
(400)(300)(200)(100)
-100200300400500
2000 2001 2002 2003 2004 2005
Total FederalMarketable Debt(12m Change)Foreign Holdings(12m Change)
Worth Watching
Growth in Credit Derivatives
Source: British Bankers' Association, US Federal Reserve, and IMF Global Financial Stability Report
Corporate Balance Sheets and Credit Swap Growth
0.0
1.0
2.0
3.0
4.0
5.0
6.0
1997 1998 1999 2000 2001 2002 2003 2004
Cre
dit D
eriv
ativ
e N
otio
nal
($Tr
illio
ns)
$(200)
$(150)
$(100)
$(50)
$-
$50
$100
$150
$200
finan
ce s
urpl
us ($
bill
ions
)credit derivatives notional
corporate f inancesurplus
The trading and notional outstanding volumes of credit derivatives market has more than doubled over the past two years. Credit default swaps represent approximately 50% of the outstanding notional, collateralized loan Obligations 22%, with asset swaps, credit linked notes, and total return swaps make up the remainder.At Present, most corporate balance sheets contain sufficient liquidity to mitigate against major credit events. The financing surplus, the difference between internal funds and investment holdings, are near record highs. Compression of that financing surplus in coming quarters, in context of slowing economy and rising interest rates suggest credit quality may slowly deteriorate.It is difficult to assess whether credit derivative markets, as well as the underlying credit market, will continue to operate smoothly in the event of a major credit event.For some reference names, some market participants perceive that the amount of protection bought or sold exceeds the value of the underlying assets. Therefore, if a credit event occurs, there may not be enough deliverable assets for all the claimants
Should Treasury Consider Reintroduction of the 30-Year
Bond?
Starting Point: Low cost borrowing over time requires issuance diversification
• Widens investor base• Provides financing flexibility• Lowers operational, event, interest risks• Facilitates efficient cash management
Questions when considering long-dated issuance
• What is optimal level of diversification?• Can we simultaneously maintain liquid
bond issuance and a short-dated bias?• Do current as well as future financing
needs and market conditions provide rationale for bond issuance?
Considerations• Effects on Portfolio Characteristics• Effects on Issuance Sizes of Existing
Securities• Effects on Costs• Effects on Refunding Needs
Percentage of Debt Maturing in Next 12 to 36 Months
25%
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 201025%
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
Maturing in 12 Months
Maturing in 24 Months
Maturing in 36 Months
Projected
Bonds are includedat $10B per semiannualauction
Projections based on OMB's FY06 Budget (except internal Treasury estimate used for FY05) and assumes coupon auction sizes remain at most recently announced amounts. Residual amounts financed with bills.
Distribution of Marketable Debt Outstanding by Security
0%
5%
10%
15%
20%
25%
30%
35%
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 20100%
5%
10%
15%
20%
25%
30%
35%
BILLS 2-3 YR NOTES 4-7 YR NOTES 8-10 YR NOTES BONDS TIPS
Projected
Projections based on OMB's FY06 Budget (except internal Treasury estimate used for FY05) and assumes coupon auction sizes remain at most recently announced amounts. Residual amounts financed with bills.
Distribution of Marketable Debt Outstanding by Security
0%
5%
10%
15%
20%
25%
30%
35%
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 20100%
5%
10%
15%
20%
25%
30%
35%
BILLS 2-3 YR NOTES 4-7 YR NOTES 8-10 YR NOTES BONDS TIPS
Projected
Projections based on OMB's FY06 Budget (except internal Treasury estimate used for FY05) and assumes coupon auction sizes remain at most recently announced amounts. Residual amounts financed with bills.
Bonds are included at $10B per semiannual auction
DEBT MATURITY MEASURES**
20
30
40
50
60
70
80
90
100
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
months
20
30
40
50
60
70
80
90
100months
Average Maturity of Issuance
Average Maturity
Includes 30-yr issuance of $10B per semi-annualauction
Projected
Projections based on OMB's FY06 Budget (except internal Treasury estimate used for FY05) and assumes coupon auction sizes remain at most recently announced amounts. Residual amounts financed with bills. **Based on end-of-fiscal-year data.
Implications of the Bond for Gross Financing Needs
68%63%
12%12%
WithoutWith Bond
Pessimistic
24%20%
7%6%
WithoutWith Bond
Central
-21%-24%
1%0%
WithoutWith Bond
Optimistic
20102006Deficit Forecast
Average change in auction size needed to meet gross marketable borrowing needs
Central Forecast based on OMB FY06 Budget (except internal Treasury estimate used for FY05), Optimistic and Pessimistic Forecasts based on OMB’s forecast errors from 1984-2004. Distribution of issuance held constant in percentage terms at FY05 level. W/ bond scenario assumes reduced bill issuance to allow for bond issuance.
Interest Cost Comparison between Current Issuance versus Bonds Included Issuance
100
150
200
250
300
350
400
450
2005 2006 2007 2008 2009 2010
Total Cost $B
100
150
200
250
300
350
400
450
Current Issuance
With Bonds
Baseline Budget and Rates
Pessimistic Budgetand Rates
Optimistic Budget and Rates
Projections based on OMB's FY06 Budget (except internal Treasury estimate used for FY05) . Assumes auction size as percent remains at the most current level. Bonds are included at $10B per semiannual auction starting in FY05. The baseline rate spread between the 30-year to the 10-year during 2005-2010 averages to about 35bps. Optimistic and pessimistic budget numbers are derived from OMB's forecast errors. Optimistic and pessimistic rates are distributions around OMB's forecasts, derived from 1962-2004 data.