Goldman Sachs Now Expects Q1 GDP of -7%

US Daily: Sharper Near-Term Correction Leads to 10% Unemployment by Year-end 2010 (McKelvey)
5:58 PM Mon Mar 2 2009

We have marked down our forecast for US economic activity in the first half of 2009; we now expect real GDP to fall at annual rates of 7% this quarter and 3% next quarter (versus declines of 4½% and 1% previously).
· The good news is that the bulk of this change is in business investment, which typically lags other sectors of the economy. Meanwhile, the steepest decline in consumer spending appears to be behind us. However, since this is already built into our forecast for remaining quarters, no further adjustments to our GDP outlook appear warranted at this time.
· As a result of the additional near-term weakness, we have boosted our expected path of the unemployment rate, to 9½% by year-end 2009 and 10% by year-end 2010; both figures are ½ point above the previous forecast. We have also marked down our estimates for the year-to-year change in consumer prices, to 1% by year-end 2009 (from 1.2% previously) and zero by year-end 2010 (from 0.5% previously).

As we noted in last Friday’s US Economics Analyst, the risks to our expectations for near-term US economic activity swung sharply to the downside last week as a result of four developments: (1) a larger-than-expected downward revision to last quarter’s setback, featuring markdowns to several components of domestic demand including a big one (0.8 percentage points) to the annualized decline in real consumer spending; (2) fresh signs of weakness in industrial activity, notably in orders for and shipments of nondefense capital goods; (3) further evidence of a sharp correction in housing activity; and (4) ongoing deterioration in the labor market, implying that the negative multiplier remains in full force. However, with several key reports due for release this morning, we held off making revisions until today.

Although these reports rendered a split verdict, downward revisions are still warranted to our expectations for US economic activity in the first and second quarters. Specifically, we now think real GDP will fall at a 7% annual rate this quarter and at a 3% annual rate next quarter. These figures compare to -4½% and -1% in our previous forecast. Quarterly details of this and other adjustments, also described in this note, are shown in the exhibit at the end of this comment.

As large as these changes are, they could have been worse if not for the mixed nature of this morning’s data. The saving grace was the 0.4% increase in real consumer spending reported for January. Although this could disappear on revision, it is the second gain in three months. The uptick in November, while only about half as large as first reported, remains reasonably sizable at 0.3% (3.6% at an annual rate). More significantly from a forward-looking perspective, consumers will begin to see some benefits from the recently enacted fiscal package, which should help ease the tight budget constraints imposed by the labor market deterioration. In addition, the tightening in credit standards during the second half of 2008, which probably played a significant role in the contraction in consumer spending, has since eased.

Thus, we take the better-than-expected data on consumer spending as a promising sign that the worst declines are over in this sector—we now think real spending will be roughly flat this quarter and start edging up next quarter, a bit earlier than before. In this regard, it is noteworthy that the saving rate rose to 5.0% in January from an upward-revised 3.9% level in December. Although this jump reflected two nonrecurring special adjustments to disposable income in January, both of them (a large cost-of-living increase in social security benefits and an even larger reduction in estimated personal income tax liabilities) represent genuine, if one-off, increases in after-tax income. Moreover, in coming months they will be replaced by additional boosts from the fiscal package, as already noted. For households who are determined to raise their saving rates quickly, these boosts will help them reach their goals with less parsimony than would otherwise have been required.

On the negative side of the data ledger, the latest report on construction outlays was much worse than expected, both in terms of the change reported for January (-3.3% versus our lower-than-average -2.0% call) and because of large revisions to November and December (2.3 and 1.0 percentage points, respectively). Although these data largely get ignored by market participants because they usually come out alongside the ISM report, they are quite important for pegging the trend in the construction components of real GDP. Given the depth and widespread nature of these declines across (residential, business, and government) subsectors of the construction industry, they overrode the positive implications of the better-than-expected consumer spending figures. (The ISM report, while also better than expected, implied an ongoing significant decline in US manufacturing and in any event does not figure directly into the GDP “bean count.”)

On balance, the pattern of recent data surprises and their impact on our economic outlook is encouraging in the sense that our forecast adjustments are concentrated in sectors that tend to lag economic activity. Thus, we have marked down estimates for real business fixed investment and real inventory investment, taking on board the unexpected weakness in capital goods orders and shipments as well as today’s construction data. Meanwhile, we have brought forward by one quarter the expected stabilization and subsequent slow improvement in real consumer spending (into the current and next quarters, respectively). The exception is housing, which appears to be even a larger drag in the current quarter than we had anticipated. However, given the low levels to which starts have fallen, we continue to think that homebuilding will stabilize later this year.

Unfortunately, our GDP revisions have negative implications for both unemployment and inflation. Specifically, we have boosted the profile for the unemployment rate to 9½% as of the end of 2009 and 10% as of the end of 2010. Both figures are ½ percentage point higher than before. (For the technically minded readers, our cumulative revision to GDP is about 1 percentage point as of the second quarter, which by Okun’s Law should be associated with an unemployment that is about ½ point higher). On the inflation front, we now expect the core CPI to rise only 1% over the four quarters of 2009 and to be flat, on balance, over the four quarters of 2010; these figures are ¼ and ½ percentage point lower than we previously thought.

With one critical caveat, the silver lining in these clouds is that unemployment and inflation are generally also lagging components of economic activity. (Unemployment leads at business cycle peaks but lags at cycle troughs.) The caveat is that the forecast revision underscores the likelihood that the US economy will endure a bout of at least mild deflation in consumer prices, if not in 2010 then in the period immediately beyond that. This is fundamentally different than the “technical” deflation we expect this year in the headline indexes, which is driven by a sharp and (so far) unsustained drop in energy prices. In principle, deflation can be self-reinforcing if price expectations are backward-looking. In this situation, consumers put off discretionary spending in anticipation of increases in real purchasing power and higher real interest rates—inevitable in markets where nominal rates are already zero—deter borrowing. However, if expectations are sufficiently forward-looking, then economic policies that are highly expansionary should head off such behavior, especially if the economy is showing signs of recovery.

Ed McKelvey

Exhibit 1: Key Numbers in the US Economic Outlook