Revised August 14, 2002

WHAT OMB’S MID-SESSION REVIEW TELLS US — AND WHAT IT OBSCURES
by Richard Kogan and Robert Greenstein

PDF of full report

Related Reports

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

In February 2001, the Administration, through the Office of Management and Budget, projected that:

Like OMB, the Congressional Budget Office also projected a $5.6 trillion surplus over the period 2002-2011.  While CBO expressed great uncertainty about its projections, however, and emphasized the wide range of possible outcomes, the Administration seemed more certain.  OMB director Mitchell Daniels stated at the time: “There is vastly more than enough room [for the tax cut]…the budget is built on very conservative and cautious assumptions.”  Daniels and OMB dismissed concerns that due to the inherent uncertainty of budget projections, surpluses of the magnitude they were projecting might not materialize.  They said that the greatest risk to the accuracy of their forecasts was that they might be understating government revenues, not overstating them.  Their views were reflected in a ringing declaration in the President's February 2001 budget: “There is ample room in the Administration’s budget to pay off debt as far as possible, to reduce taxes for American families, to fund program priorities, and still have roughly $1.0 trillion [outside of the Social Security Trust Fund] for Medicare modernization and to meet other programmatic and contingency needs as they arise.”

President Bush similarly stated at the time, “We’re going to prove to the American people that we can pay down debt, fund priorities, protect Social Security, and there will be money left over, which we strongly believe ought to be passed back to taxpayers. … We can proceed with tax relief without fear of budget deficits, even if the economy softens. …”

Now, 17 months later, the OMB “mid-session review” released on July 15 shows that $3.9 trillion of the $5.6 trillion ten-year surplus has already disappeared.  Moreover, in the budget it issued this February, the Administration proposed tax cuts and spending increases totaling an additional $1.3 trillion over the ten-year period (and far more in the decade after 2011); as the mid-session review indicates, the Administration continues to support these proposals.  According to OMB’s own projections, this would leave the budget outside Social Security in deficit through 2012 (and probably for decades thereafter).

 In short, OMB’s new budget projections are dramatically different from its projections of 17 months ago.  Some of the change in the budget picture reflects increases in security spending.  Most of the change, however, is due to other factors — the large tax cut enacted last year (which the new OMB data show to have caused 38 percent of the reduction in the projected surplus over ten years), a less buoyant economy than had been forecast, and downward revisions in the forecasts of how much tax revenue the economy generates for any given level of economic performance.  It is evident that the Administration’s forecast of 17 months ago was too optimistic about the state of the economy and revenue collections, too dismissive of the impact that emergencies and unforeseen factors could have on the budget, and, as a consequence, too sanguine about the nation’s ability to address various priorities and still have enough left over for the Administration’s tax cut.

Analysis of the Administration’s new mid-session budget indicates that it suffers from some of the same types of shortcomings:

As a result, the new budget substantially overstates expected government revenues, while substantially understating expected government costs.

            The recent disclosures of misleading corporate accounting practices and rosy corporate financial projections are now taking a serious toll on the stock market and beginning to affect consumer confidence.  At this juncture, it is important for government to avoid the temptation to engage in accounting maneuvers that overstate likely revenues, understate likely expenditures, and advance proposals whose full costs are concealed by slow phase-ins or other delays in implementation.

Table 1
Shares of the Reduction in the Projected Ten-Year Surplus, 2002-2011
Changes from February 2001 baseline to July 2002 baseline

What OMB July 12 Press Release Claimed

What OMB Data Show

Recession

67%

All economic reestimates (incl. recession)

10%

Technical reestimates

no mention

Technical reestimates

33%

“Security and the war”

19%

All enacted spending increases

18%

Last year’s tax cut

15%

Last year’s tax cut

38%

Tax cuts in “stimulus” bill

no mention

Tax cuts in “stimulus” bill

2%

TOTAL

100%

TOTAL

100%

Source: OMB.  Percentages may not add to 100 due to rounding.

This analysis examines the mid-session review and the questions it raises in some detail.  It starts with an examination of factors that make the budget projections too optimistic.  It then examines the causes of the change in the budget picture since February 2001, based on OMB’s own data.  Finally, it looks at the implication in the mid-session review that Congressional efforts to increase appropriations for domestic programs constitute the principal threat to fiscal discipline.

 

A Rosy Scenario

             “Rosy Scenario” is the name given to projections the Reagan Administration issued in 1981, which showed that the large Reagan tax cuts would generate such strong economic growth that the budget would turn from deficit to surplus despite the loss of revenue from the tax cuts and the Administration’s proposed increases in defense spending.  Reality was not kind to Rosy; as a share of the economy, federal deficits were higher from 1982 through 1986 than in any year since World War II, and the 1983 deficit exceeded the peak deficits of the Great Depression.

             In retrospect, it now seems clear that President George W. Bush’s first budget also was built on what turned out to be a rosy scenario.  The new projections in the mid-session review follow such a path, as well.

Experts believe that capital gains tax revenue will drop further in fiscal year 2003.  Capital gains tax collections in fiscal year 2003 will largely reflect the capital gains income shown on 2002 tax returns, which will be filed next winter and spring.  And capital gains income for tax year 2002 is expected to fall because of the significant declines in the stock market, with the S & P 500 index now down more than 25 percent this year.  Yet OMB’s mid-session review appears to assume that capital gains revenue will increase in fiscal year 2003 above the fiscal year 2002 level.[1]  For that to occur, the stock market would generally have to produce a stunning turnaround by December 31, essentially recovering its losses for the year and then some, so that people who sell stock by then (and pay capital gains taxes on it next winter and spring) will be selling at a gain.  Given the current state of the stock market, it seems highly questionable for OMB to be counting on stock prices taking off this much in the immediate future.  This dubious assumption is a key reason that the mid-session review shows a substantial reduction in the deficit in fiscal year 2003 when most experts expect the deficit to grow larger.

            The mid-session review also understates likely future deficits by omitting the costs of a number of expensive policies that the President supports and Congress is virtually certain to enact.[2]

Because the Administration's budget omits extension of this provision, the revenue numbers in the budget are based on the assumption that the number of taxpayers subject to the AMT will swell from 1.4 million in 2001 to 39 million by 2012.  (Buried in one of the budget books is an acknowledgment that the revenue numbers assume that 39 million taxpayers — one of every three in the nation — will be subject to the AMT by 2012.)  There is no possibility the Administration or Congress will allow this to happen.  The Administration clearly intends to propose addressing this problem before the current AMT relief provision expires in 2004, and there is no question that AMT relief will pass.  OMB Director Daniels recently acknowledged in congressional testimony that the AMT problem will need to be addressed.  Joint Tax Committee estimates indicate that the cost of resolving this problem will amount to several hundred billion dollars over the next ten years.

Moreover, assuming that a swollen AMT will be in effect in 2011 and 2012, as the OMB projections do, helps the Administration in a second way — it substantially reduces the amounts that OMB shows as being the costs in 2011 and 2012 of extending the tax cuts scheduled to expire in 2010.  Since the AMT is assumed to affect 39 million taxpayers by 2012, it is assumed to cancel out a significant share of the tax cuts in those years, sharply lowering the amounts that OMB prints in the budget for the cost of extending the tax cut.

 

How the Budget Has Changed Since February 2001

             When OMB projected a $5.6 trillion, ten-year surplus in February 2001, the surplus seemed almost too big to squander.  In this section of the analysis, we use the detailed budget data that OMB released in conjunction with the mid-session review to examine how, according to OMB’s numbers, the budget outlook for the next ten years has changed since early 2001.

What OMB said.  The material that OMB distributed to the press on July 12 made striking claims about the causes of the deterioration of the surplus.  The OMB press release asserted:

“…the recession erased two-thirds of the projected ten-year surplus (FY2002-11).  The costs of security and war lowered the projections 19%.  The tax cut, which economists credit for helping the economy recover, generated less than 15% of the change.”

             OMB’s own numbers, however, show that these assertions are incorrect.  The percentages cited in the OMB press release are the figures for the two-year period 2002-2003 or for 2002 only, not for the ten-year period 2002-2011.  The figures for the ten-year period are very different.  The tax cut phases in over time, and its costs mount markedly as the decade progresses.  It accounts for a much larger share of the deterioration of the surplus over the next ten years as a whole than of the deterioration just in 2002 and 2003.[4]  (This error apparently originated with an overzealous public relations person in OMB or the White House.  The error was so flagrant, however, that OMB was clearly aware of it, and as described below, senior Administration officials were questioned about it within a few days.  Nevertheless, OMB did not drop the erroneous paragraph from its July 12 press release until July 26, after the initial release of this Center analysis calling attention to the error, and did not amend the press release to note that the original version contained errors until August 7.)

In the short run, the recession is indeed an important factor in the shrinkage of the surplus.  But over the long run, this cannot possibly be the case — OMB is not predicting a ten-year recession.[5]  (CBO previously concluded that a mild recession would decrease projected ten-year surpluses by less than $150 billion, which suggests that 90 percent or more of the $3.9 trillion shrinkage of the surplus since January 2001 is not attributable to the recession.)

What the OMB data actually show.  As noted, the projected $5.6 trillion, ten-year surplus projected by OMB in February 2001 has shrunk by $3.9 trillion.  The following tables present what the OMB data reveal to be the reasons for this deterioration.[6]

As Table 2 shows, the tax cut accounts for 16 percent of the deterioration of the surplus in 2002 and 2003, but will account for nearly half of the shrinkage of the surplus in 2010 — or nearly as much as all other factors combined.  Over the ten-year period as a whole, the tax cut accounts for 38 percent of the shrinkage of the surplus, which is more than spending increases for defense, homeland security, and domestic programs and changes in the economy combined.[7]  The tax cut is the single largest factor responsible for the deterioration of the surplus over the ten-year period.  (Moreover, the tax cut appears to be at least 2½ times as costly over the next ten years as the war on terrorism.[8])

Under tough questioning in a Congressional hearing on July 17, Glenn Hubbard, Chairman of the President's Council of Economic Advisers, admitted that the tax cut accounts for about 40 percent of the deterioration in the surplus over the next ten years and that the 15-percent figure applies only to the first year or two.[9]

Table 2
The Surplus Has Shrunk by $3.9 Trillion since February 2001

Difference between February 2001 baseline and July 2002 baseline, in billions of dollars

 

 

2002-2003 avg.

2010

2002-2011 total

 

$

%

$

%

$

%

Changes in the economic forecast

51

12%

31

7%

386

10%

Changes due to “technical” reestimates

185

45%

144

31%

1,282

33%

Legislation enacted to date:

 

 

 

 

 

 

 

Last year’s tax cut

68

16%

218

47%

1,491

38%

 

Tax cuts in stimulus bill

44

11%

1

0%

71

2%

 

Program increases

67

16%

73

16%

689

18%

Total reduction in the surplus

414

100%

467

100%

3,920

100%

Source: CBPP calculations from OMB data.  Figures may not add due to rounding.

A second way to analyze the OMB data is to distinguish those matters over which policymakers have little control — the economy and changes in “technical factors” that affect revenue collections and the costs of certain programs such as Medicare — from changes in the surplus caused by legislation that Congress passed and the President signed.  Congress and the President should not be held responsible for changes in the budget that result from unanticipated changes in the performance of the economy or from technical reestimates of taxes or expenditures, although they can be charged with adopting too rosy a scenario to begin with.  Congress and the President are responsible for the legislation they enact.

According to OMB, legislation has reduced the projected 2002-2011 surplus by $2.3 trillion since February 2001 and accounts for the majority of the $3.9 trillion surplus shrinkage.  Examination of the OMB data shows that this $2.3 trillion deterioration was due predominately to the tax cut, as shown in Table 3, below.

Table 3
Legislation Has Shrunk the Surplus by $2.3 Trillion since February 2001

In billions of dollars

 

2002-2003 avg.

2010

2002-2011 total

 

$

%

$

%

$

%

Last year’s tax cut

68

38%

218

75%

1,491

66%

Tax cuts in stimulus bill

44

25%

1

0%

71

3%

Program increases enacted to date

67

38%

73

25%

689

31%

Total cost of legislation to date

179

100%

292

100%

2,251

100%

Source: CBPP calculations from OMB data.  Figures may not add due to rounding.

Still another way to analyze the OMB data is to assess what portion of the surplus deterioration is due to the shrinkage in revenues (including both the shrinkage due to tax cuts and the shrinkage due to economic and technical factors), what proportion is due to increased spending on defense and homeland security, and what portion is due to domestic spending increases.

            These figures, shown in Table 4, include the increase in interest payments caused by reduced revenues and higher spending (see footnote 6).  Over the ten-year period from 2002 through 2011, OMB now projects that interest payments on the debt will be about $950 billion higher than it projected in February 2001.

Table 4
Combined Effect of Legislation and Reestimates on Projected Surpluses:
Change in OMB Baseline Projections from February 2001 to July 2002

In billions of dollars

 

2002-2003 avg.

2010

2002-2011 total

Reduced revenues

77%

95%

87%

Increased costs of defense and homeland security

9%

6%

8%

Increased costs of domestic and international programs

14%

-2%

5%

Total reduction in the surplus

100%

100%

100%

Source: CBPP calculations from OMB data.  Figures include the extra interest payments due to each of the causes listed.  Figures may not add due to rounding.

            These figures cast doubt on the statements made in early 2001 that the federal government had so much excess revenue that it could nearly eliminate the national debt, increase spending on defense, education and other needs, provide a prescription drug benefit, enact the Administration’s entire tax cut, and still have a cushion of close to $1 trillion left over for contingencies.

 

Misleading Analysis of Appropriations

            Some OMB material accompanying or contained in the mid-session review appears designed to exert pressure on Congress to limit the level of annual appropriations.  According to OMB, the programs whose funding is determined through annual appropriations (known as “discretionary” programs) are the main threat to future surpluses.  On page 9 of the mid-session review, OMB purports to show that the President has requested a hefty 10 percent funding increase for these programs for 2003 (primarily for defense and homeland security) but that the budget plan the Democratic majority of the Senate Budget Committee approved in April contains a 12.3 percent funding increase even while it scales back the President’s defense request.  According to the OMB presentation, the Senate Budget Committee plan includes large increases in domestic programs.

            OMB’s portrayal of the Administration’s proposed discretionary funding increases is misleading, and its portrayal of the spending reflected in the Senate Budget Committee plan is incorrect.  Table 5 shows the levels of discretionary funding (or appropriations) requested by the President for fiscal year 2003, as well as the comparable levels included in the congressional budget resolution approved by the Senate Budget Committee in April.  This table clarifies a number of issues and shows that the mid-session review overstates the size of both the funding increases that the Administration is proposing for discretionary programs and the increases contained in the Senate Budget Committee plan.

          The overstatements reflect a curious maneuver on OMB’s part.  In calculating the percentage increases in discretionary funding both in its budget and in the Senate plan, OMB has omitted from the 2002 levels the funding that was enacted last fall in response to the terrorist attacks.  But in calculating the discretionary funding levels that would be provided for fiscal year 2003 under both the Administration’s budget and the Senate Budget Committee plan, OMB has counted the continuation of this anti-terrorism funding.  By comparing 2002 funding levels that omit anti-terrorism funding with proposed 2003 levels that include such funding, OMB has artificially inflated the increases between 2002 and 2003.

             In addition, OMB claims in the mid-session review and accompanying press materials that the Senate Budget Committee plan would reduce the President’s 2003 defense request.  In fact, the Senate plan includes the full amount the President requested for 2003 for defense (and for homeland security).

             Finally, while the overall discretionary funding level in the Senate plan does exceed that in the Administration’s budget — with virtually all of the difference being due to the higher levels the Senate plan contains for domestic discretionary programs such as education, veterans’ medical care, and natural resources — the Senate’s so-called “increase” in discretionary funding turns out simply to be a smaller reduction in domestic discretionary programs than the Administration has proposed.  The mid-session review gives the impression that the Administration has proposed adequate but restrained increases for domestic discretionary programs while the Senate Budget Committee plan calls for unsustainably large jumps in such funding.  As Table 5 shows, the Administration has, in fact, proposed to reduce overall fiscal year 2003 funding for domestic discretionary programs outside of homeland security by $15 billion below the CBO baseline (i.e., $15 billion below the 2002 level adjusted for inflation). 

Table 5
Funding For Discretionary Appropriations, 2002 and 2003
[10]
In billions of dollars

 

 

Defense, homeland security, and international

Domestic (except homeland security)

Total

FY 2002 funding level enacted last year

           $389

$321

$710

FY 2003 funding levels:

 

 

 

 

CBO baseline (2002 adj. for inflation)

             400

333

732

 

President’s request (CBO est.)

             442

318

759

 

Senate Budget Committee

             442

326

768

 

 

 

 

 

 

 

Increase or decrease, 2002 to 2003:

$

%

$

%

$

%

 

President’s request (CBO est.)

+52

+13.4%

-3

-1.1%

+49

+6.9%

 

Senate Budget Committee

+53

+13.5%

+5

+1.6%

+58

+8.1%

Increase or decrease from CBO’s baseline:

 

 

 

 

 

 

 

President’s request (CBO est.)

+42

+10.5%

-15

-4.6%

+27

+3.7%

 

Senate Budget Committee

+43

+10.6%

-7

-2.0%

+36

+4.9%

  Source: CBO.  Figures may not add due to rounding.

            The Senate Budget Committee plan would set overall funding for domestic discretionary programs other than homeland security some $7 billion below the CBO baseline level.  The current dispute between the Administration and the Senate thus essentially concerns not domestic funding increases but rather the depth of domestic funding cuts.  It is not valid to imply that smaller cuts (which the Administration portrays as larger increases) pose a serious threat to future budget discipline.  That threat lies elsewhere.

            To be sure, total funding for discretionary programs — including the defense budget and the homeland security initiatives — would rise significantly under both the Administration’s budget and the Senate Budget Committee plan.  But this is entirely because of the large defense and homeland security increases that both budget plans contain.


End Notes:

[1]   Two elements of the mid-session review suggest that OMB expects a noticeable bounce-back in capital gains revenues in 2003.  The first is that OMB forecasts individual income tax revenues to grow by 10.4 percent from 2002 to 2003 even while it forecasts the economy to grow by 5.5 percent.  (These figures include both inflation and real growth).  Revenue growth exceeds economic growth by such large margins only when the vast bulk of income growth is concentrated at the top of the income scale, as occurs when the stock market booms.
The second element is buried in OMB backup material that shows how the revenue forecast has changed since February 2002.  In that material, OMB shows a permanent “technical” decrease in the amount of revenues it now projects, compared with its projection of February 2002.  “Technical” changes in budget estimates are changes that are not caused by legislation and not directly related to the level of growth in the overall economy.  A lower projection of capital gains revenues is a prime example of such a technical change, and OMB has expended considerable ink explaining (correctly) that capital gains income cannot grow as it did in the past if the stock market is flat or falling.  Given the stock market’s performance so far this year, one would expect to see large “technical” reductions in the revenue forecast for 2002 and even larger reductions for 2003 (since capital gains revenues in any fiscal year are paid on capital gains realized in the prior calendar year), relative to the projection OMB issued in February 2002.  But OMB shows a smaller downward technical reestimate for 2003 than it does for 2002, implying a big bounce-back in the stock market by the end of this year.

[2]   See also President's Budget Uses Accounting Devices And Implausible Assumptions to Hide Hundreds of Billions of Dollars in Costs,” Center on Budget, February 5, 2002, available at http://www.cbpp.org/2-4-02bud.htm. 

[3]   “Daniels Repeats Bush Veto Threat on Approps,” Roll Call, July 18, 2002, p 3.

[4] The statement that the tax cut is responsible for only 15 percent of the deterioration in the surplus is also misleading in two other ways.  First, the 15-percent figure excludes the costs of the corporate tax cuts enacted in the stimulus bill.  Second, OMB calculated the 15-percent figure as a share of the total change from the February 2001 baseline through the (assumed) enactment of the President's 2003 budget.  Under this approach, the additional tax cuts proposed by the President and his requested 2002 supplemental appropriations bill and 2003 defense increases are used to make the deterioration attributable to last year’s tax cut appear smaller.  Under this method, the bigger the new tax cuts and defense increases that the President requests, the smaller last year’s tax cut appears as a share of the total deterioration.

[5] The statement in the OMB press release that the recession caused two-thirds of the deterioration of the surplus is problematic for a second reason as well: not only does it apply only to 2002, but it also mixes together temporary revenue shortfalls caused by the recession with temporary and permanent revenue shortfalls stemming from such reasons as the overvalued stock market, which cannot be attributed simply to the effects of a temporary recession.

[6] Projections of the surplus or deficit can deteriorate for a specific reason (e.g., a revenue shortfall, a tax cut, or program increases).  Each such change also produces a change in the amount of interest projected to be paid on the federal debt.  For instance, OMB estimates that last year’s tax cut directly cost $1.2 trillion over the ten-year period 2002-2011; this means that the surplus will be $1.2 trillion smaller than projected in February 2001 for that reason alone.  The smaller surplus means that federal debt will be higher in each year than was projected in February 2001, and the Treasury consequently must pay more interest on the debt than was projected.  OMB estimates that last year’s tax cut will cause the Treasury to pay $305 billion more interest on the debt over the ten-year period, bringing the total ten-year cost of the tax cut to $1.5 trillion.  OMB’s approach of attributing interest costs to legislation that directly costs money is sound and is followed in the tables in this analysis.

[7]   An analysis issued by the Senate Budget Committee majority staff similarly finds that 38 percent of the reduction in the ten-year surplus is attributable to last year’s tax cut, using the same comparison of OMB data as we provide in this analysis.  The House Budget Committee minority staff makes a different comparison.  Rather than comparing OMB’s February 2001 baseline with OMB’s new baseline, the HBC minority compares OMB’s February 2001 baseline with the President’s current budget proposal.  Because that budget calls for an additional $1.3 trillion in costs over the ten-year period 2002-2011, last year’s tax cut is a smaller percentage of the total possible deterioration in the surplus.  Specifically, the $1.5 trillion cost of the tax cut over 2002-2011 is 29 percent of a $5.2 trillion deterioration.  But, as the House Budget Committee minority staff points out, the President’s budget proposes significant additional tax cuts (and also reflects the tax cuts enacted this spring in the “stimulus” bill), and these tax cuts should be taken into account as well.  In total, OMB numbers show that enacted and proposed tax cuts constitute 37 percent of the total $5.2 trillion deterioration that will occur if the President’s budget is enacted by Congress. 

[8]   The House Budget Committee minority estimates that the ten-year cost of the war on terrorism is likely to be less than $600 billion, including interest.  See http://www.house.gov/budget_democrats/analyses/2003_midsession/msr_memo.htm.

[9] “Tax Cut ‘Bit Player’ in Revised Deficit Forecast, CEA Chairman Says,” Daily Tax Report, Bureau of National Affairs, July 18, 2002, p. G-8.  The Daily Tax Report article quotes Hubbard as saying, “In the first year, 15 percent sounds about right. … Over the longer period of time, a 40 percent number sounds about right.”

[10]   The figures shown in this table are CBO’s March estimates of the President’s February budget request.  CBO has not undertaken an estimate of the President’s mid-session review, but the mid-session review includes only tiny changes to the President’s budget request for discretionary 2003 funding, so CBO’s March figures continue to apply. 
The President’s February budget omitted any request for supplemental appropriations for 2002 even though OMB was at work preparing such a request at that time.  While the President’s budget now reflects his request for a 2002 supplemental appropriations bill, we do not include it in CBO’s 2002 figures in Table 5 because this table is intended to compare the President’s 2003 request with the 2002 amount enacted last year.  If we had included the pending 2002 supplemental funding in the base 2002 level, the increase from 2002 to 2003 — which Table 5 shows is smaller than OMB asserts — would appear smaller still.