From Chowdhury and Huie, "Skyrocketing Issuance," US Economics/Strategy Weekly (Deutsche Bank, 10 Oct.) (not online):

Treasury issuance is likely to increase to extraordinary levels over the past year. There are 3 components to the issuance picture. The first is the traditional federal budget, which in fiscal year 2009 is likely to increase substantially from the 2008 deficit of around $440 bn. The second are the various Treasury rescue initiatives that involve buying assets or equities; only the expected net cost will be formally recorded on the budget, but the entire gross spending amount will be added to the issuance requirement. Finally, the Federal Reserve's liquidity facilities will also add to issuance, as the Fed no longer has capacity to sell or lend the Treasuries in its portfolio; instead, going forward it will rely on the Supplementary Financing Program, where the Treasury issues bills and deposits the proceeds at the Fed, to finance its lending facilities. In total, we expect net issuance to rise to $3.3 tn over the fiscal year.

Federal budget

For the on-budget contribution to issuance, we are expecting a $775 bn deficit in FY 2009. This is based on the CBO baseline, adjusted for increasing baseline expenses, falling revenue (particularly from corporate income taxes), potentially large fiscal initiatives, and a fiscal stimulus package from the new Administration that would all add up to a near doubling of the traditional measure of the budget deficit. For our estimate, we are assuming $100 bn outlays for FDIC rescues and as a fiscal stimulus, as well as $35 bn above the baseline for purchases of GSE preferred stock. The actual outcome relative to our budget deficit estimate is biased upward, since there could easily be a larger fiscal package, the FDIC outlays could move sharply higher if more banks fail, and tax revenues could fall if the economy enters into a deep recession.

I illustrate where that puts the budget deficit (not net debt issuance) for FY2009, in the context of the CBO's baseline, assuming the 2001 and 2003 tax cuts are made permanent (or alternatively if the cuts are re-arranged, but retain the same budget deficit implications -- see this post).

Figure 1: Budget balance to GDP ratio (blue), baseline (red), EGTRRA/JGTRRA extended (green), and Deutsche Bank estimate for FY 2009 (blue square). Source: Author's calculations based upon CBO, The Budget and Economic Outlook: An Update (September 2008)Table C-2 and Table 1-8 [xls], Deutsche Bank, "Skyrocketing Issuance," and author's calculations.

This means the Bush Administration will bequeath to the next Administration a budget deficit almost as large as that experienced under the Reagan Administration.

(Note: Net debt issuance differs from the budget deficit since the proceeds from debt issuance will be used to purchase assets. See the CBO director's discussion here.)

The report also notes that the maturity of the new issues will be skewed toward short end -- at least the portion not having to do with the budget deficit, largely because of the demands of foreign investors:

...A large proportion of the Treasury issuance will likely be bought by foreign investors, who prefer to buy short maturities. These investors are focused mainly on the currency exposure -- choosing dollar assets versus euro or other currencies -- and less on the yield of the asset. ...

So, the fear that we would be dependent upon foreign capital exactly at the time it was hardest to get capital has come to pass, although not exactly in way I expected. Here, it is likely that we can obtain the financing, but at shorter maturities than we would like otherwise. And excessive reliance on short term debt has, in other circumstances, led to vulnerabilities to capital reversals.

I am compelled to thinking about what could have been. As I wrote about two years ago:

...By virtue of the fiscal and monetary policies of the last five years, the U.S. is ever more dependent on foreign capital inflows to determine interest rates; and indeed the United States is as dependent as it has ever been (at least in the post-War era) on foreign official or quasi-state -- not private investor -- financing. How all the ties binding the world's single largest economy will be unwound is something nobody can be certain of. What we can be certain of is that the choices made by policy makers in the past five years have circumscribed our ability to manage a downturn.

And that was, of course, before the $240-300 odd billions spent in Iraq over the last two fiscal years alone [0]. That being said, there are now many calls for fiscal stimulus in the face of the possibility of a deep and prolonged recession, despite the yawning deficit (contingent liabilities strike!). One particularly persuasive argument is made by Barry Eichengreen; given his knowledge of history, I think his recommendations should be given close attention (although I'm not certain I'd go with 5 ppts. of GDP stimulus, given portfolio balance effects). But the situation does look pretty dire -- as indicated by the WSJ survey of forecasters.

Figure 2: Output gap, calculated as log deviation from potential GDP, from 26 Sep release (blue line), and implied GDP gaps from WSJ survey (October survey, red line) and highest for 2008-09 from MacroEcon Global Advisers (teal), and lowest from MFR (green). Source: BEA, CBO [xls], WSJ [xls], and author's calculations.

(Note that this figure differs from Figure 2 in this post in that the "low" forecast was for the 2008Q3-09Q2 period, while this "low" forecast is for the 2008Q4-09Q4 period; the "hi" forecast is MacroEcon Global Advisers in both instances, though).

The mean forecast is for output gap in excess of 4% (log terms) by 2009Q4. While not unheard of, there remains considerable downside risk. The low-growth forecast implies a substantial 6% gap (albeit one less than the 8% gap in 1982Q4). The high estimate by MacroEcon Global Advisers indicates only about a 1% negative output gap, but as one can see from the relation of this forecast to the mean forecast, it's an outlier. This forecaster's forecast for the 2008Q3-09Q4 period is about 4 standard deviations (!) from the mean.

Figure 3: Histogram for 2008Q3-2009Q2 growth rates. Source: WSJ [xls].

If we were to undertake further fiscal stimulus, how should we structure it? I would say the way not to proceed is by enhanced tax cuts to the highest income quintiles. The marginal propensity to consume is highest for the liquidity constrained invidividuals, so that would argue for skewing any additional tax cuts to lower income households, if tax cuts are to be pursued at all.

We really need the biggest bang for the buck, given that we have borrowed so much already. That should be an organizing principle for any fiscal policy initiative, and I think is congruent with EconomistMom's admonition to not abandon fiscal responsibility even in these dire straits. In a macro context, transfers (such as unemployment insurance) will also have a greater impact for the same reasons as why tax cuts aimed at lower income groups have biggest multiplier effects. Counterarguments that extended unemployment insurance payments will only be saved make sense in a permanent income context; but if liquidity constraints are binding, recipients are more likely to spend a higher proportion.

As anybody who's taken intermediate macro knows, in a Keynesian framework (arguably the relevant one when there are underutilized resources in the economy, i.e., when the aggregate supply curve is pretty flat), a dollar's worth of expenditures on goods and services yields a larger multiplier than that on transfers or tax cuts. This suggests that supporting already planned investment in infrastructure -- by states and municipalities that are currently liquidity constrained [1], and squeezed by declining tax revenues -- would be a more effective way of supporting aggregate demand than tax cuts (state and local balance on a NIPA basis has decreased by about 1 ppt of GDP since 2006Q1, suggesting that there will be further compression in state and local spending as the balanced budget requirements, credit constraints, and declining tax receipts collide)[2]. That suggests aid to states as a means of fast spending (the argument against discretionary countercyclical fiscal policy is usually that the outside lag is too long).

Many of these arguments will look familiar. I laid out many of these same points when discussing the first stimulus plan [3] [4], back in January.

Still, too bad we didn't get to a zero structural budget balance (last seen in 2000) [5] when we had the chance.

Figure 1 from CBO, The Cyclically Adjusted and Standardized Budget Measures, October 2008. "The cyclically adjusted deficit or surplus attempts to filter out the effects of the business cycle. The standardized budget measure removes the effects of other factors in addition to those of the business cycle. Potential gross domestic product is the level of output that corresponds to a high level of resource—labor and capital-use. The shaded vertical bars indicate periods of recession. (A recession extends from the peak of a business cycle to its trough.) The dashed vertical line separates actual from projected data."

Note: The CBO projections assume the EGTRRA/JGTRRA provisions expire, as established in the original legislation.

A comprehensive survey of the effectiveness of discretionary, counter-cyclical fiscal policy is in Chapter 5 of the IMF's World Economic Outlook. [pdf]

Technorati Tags: recession, budget deficit, fiscal policy, federal debt, and infrastructure.