This WSJ article makes it complicated but is quite conventional:
Whether the rest of the year is better or worse depends on four factors: Will consumers spend more readily? Will business investment pick up? Will federal, state and local governments continue to retrench? And will U.S. exporters manage to sell more goods and services to the world?
This year, the four key GDP components—consumption, investment, government and trade—have all fallen short of forecasts. But economists, citing factors such as lower gasoline prices and easing supply disruptions from Japan, say growth will be stronger in the second half. To set the recovery back on course, some or all of the four need to do better.
And goes on to waste a lot of ink trying to figure how each component of GDP (C+I+G+(X-M)) will fare going forward.
One can make it much simpler. Whether or not the rest of the year will be better or worse depends on just one factor: Whether Monetary Policy will be geared to drive nominal spending or not. Alternatively, will the Fed get up from its ring-side seat and play ball?
And no, Q3 won´t win the game because the opposing team already figured how the “strategy” works.
And forget Fiscal Stimulus. It has done much harm already.
One of the more interesting “contributions to GDP analysis” comes from the NX (net exports) component. Often we read: “Imports depressed growth” or, “If it weren´t for the rise in imports GDP growth would have been higher”.
But when asked “Why did imports grow so much”? The answer usually is “Because economic growth was strong”!
There´s no clearer way to show that “component analysis” is not illuminating!
From Scott Sumner in one of his first posts back in February 2009:
Macroeconomics should be about aggregates, not components of spending. Yes, changes occurring in the various components of GDP can impact interest rates, and thus velocity. And if monetary policy is inept (i.e. doesn’t offset changes in velocity) that can impact nominal spending, but it certainly isn’t the most illuminating way of looking at the issue. It’s like trying to explain changes in the overall price level by modelling changes in the nominal price of each good—theoretically possible, but a waste of time.