Bill Gross of PIMCO recently wrote a commentary on the rising amount of public and private debt in the US. He argues that the US's total debt cannot keep increasing as a percentage of GDP, and that Very Bad Things will occur when it stops increasing.

To take Arjun's example, say lots of Chinese feel that China is a better place to invest. This will cause the following chain of events, ceteris paribus:

  1. Demand for US assets falls, so there is less investment in the US.
  2. Demand for dollars falls, so the dollar weakens.
We know that:
 GDP = C + I + G + NX
  C  = Consumption (people and corporations)
  I  = Investment
  G  = Government Expenditures
  NX = Net Exports (Exports - Imports)
If the dollar weakens, NX will improve. On the other hand, I will fall due to less overall investment. So I will have to fall by more than the gain in NX. But we know that if China refuses to take dollars, China will insist on imports equaling exports, so it can trade as much as possible without taking dollars.

If this is the case, the loss in I from China pulling out will equal the gain in NX, as we just erased the trade deficit between China and the US (the Chinese surplus of which was invested back into the US).

Now, we need to think about what happened to the interest rate. Since less people are demanding US bonds, the rates for these have increased. This will force up interest rates in general in the US. This will cause Investment to fall, as a bunch of projects that were profitable just got edged out by the higher interest rate. This means C falls, causing GDP to fall overall.

Now we get to the crux of Gross's argument. The US uses increasing amounts of debt to move its GDP forward. If access to debt is restricted, the economy might be leveraged into reverse.

If GDP is falling because I is falling, consumers will be more apt to save, so C will fall also. In the past, this downward spiral was countered by increases in G, governmental spending (Classic Keynesian). But if the government already has a bunch of debt outstanding, attempts to sell more bonds will only increase the general interest rate level. And this would be a Very Bad Thing, as the bulwark of last resort (the US Government) was hindered by its current debt.

Gross essentially shows why deficits are bad unless they are truly needed, but his argumentation is extremely elliptical. The private sector can get crazy with loans, that's bound to happen, but the Government must be able to supply the spending when the private sector loses it.