The New Economist -- one of my new favorites in econblog world -- links to a Federal Reserve Board working paper by Hilary Croke, Steve Kamin, and Sylvain Leduc suggesting that soft landings appear to be the norm.  As befits any good work, they first define what they mean by "soft landing."

... in our analysis, we will take the disorderly correction hypothesis to refer to a scenario in which exchange rate depreciation prompts increases in interest rates and declines in stock prices which ultimately push the economy into recession, even as financial markets continue to operate smoothly.

Here's what they find.

... we found little evidence among past adjustment episodes of the features highlighted by the disorderly correction hypothesis. Although the contraction episodes in our sample experienced significant shortfalls in GDP growth, these shortfalls were not associated with significant and sustained depreciations of real (price-adjusted) exchange rates, increases in real interest rates, or declines in real stock prices. By contrast, it was among the expansion episodes, where GDP growth picked up after the onset of adjustment, that the most substantial depreciations of real exchange rates occurred. These findings certainly do not preclude the possibility that future current account adjustments could be disruptive, but they do weaken the historical basis for predicting such an outcome.

Here's a brief summary of the mean experience across the 23 episodes they study.

The median current account deficit peaks at around 4 percent of GDP, and narrows to near 1 percent of GDP after two years...

GDP growth appears to decline well before current account adjustment begins, and then to bottom out at a low, but positive, level–about 3 percentage points below its initial level–a year or so after the current account balance reaches its trough. This is consistent with the view that current account adjustment may entail some cost in terms of GDP growth, but not a severe disruption to economic activity...

There is no evidence of a sustained inflationary surge during current account adjustment episodes. Headline CPI inflation picks up a bit during the first full year of adjustment, but then moves down, ending up several percentage points below its initial value...

The real effective exchange rate begins to depreciate... a bit before current account adjustment begins. Interestingly, total real exchange rate depreciation appears to average only about 8 percent, which seems quite small compared with the substantial average current account adjustment described above...

[The long-term real interest] rate appears to have remained reasonably stable over the adjustment period, moving up a bit prior to the onset of adjustment and down a bit thereafter...

Both saving and investment rates decline from two years before to two years after adjustment beings, but investment declines a bit more. Recall that the current account balance is essentially equal to saving minus investment. Most of the decline in saving rates takes place before adjustment, thus leading the current account balance to deteriorate. By contrast, investment rates start declining after the onset of adjustment, this leading to the correction of current account deficits.

The authors do note that

... the U.S. current account deficit, at present, is already larger than it was for the average of the episodes in our sample... On the one hand, this could mean greater dislocations in financial markets and economic activity during the adjustment process.

But

On the other hand, as noted above, we found no evidence that episodes with greater degrees of current account adjustment also experienced larger shortfalls in GDP growth.

This is a very accessible paper, and well worth a look if you are interested in this subject.

UPDATE: Sure, but does it work in theory?  (Hat tip, Calculated Risk.)