The traditional explanation is that there are lags in the market for wages, for equilibrium aggregate supply and demand (if aggregate demand can even be fully conceptualized), and expectations.

Real business cycles say that even when markets clear there can be changes in the total economy.  The biggest of these changes is that the marginal productivity of labor somehow drops (this can be the removal of complimentary capital investment, or at least slowing in the replacement rate).  Higher than market interest rates could slow the investment in capital, allowing capital stock to depreciate, reducing the absolute total. 

“Many economists find the real-business cycle theory totally unbelievable. No one can observe the technological shocks that are at the heart of this explanation, and it strikes many as simply ridiculous to argue that the unemployment during a recession is voluntary. On the other hand, the economists who have formed these arguments are among the brightest of the profession, and they can show that the patterns that their mathematical models generate are remarkably similar to the patterns that the real world generates.”  -- http://ingrimayne.saintjoe.edu/econ/Connections/Real-Business.html 

From the archive, Tyler’s previous comments:

http://www.marginalrevolution.com/marginalrevolution/2004/10/why_real_busine.html 

Note:  Capital is used to produce consumption goods (or other capital goods whose final end purpose is consumption goods, the production process eventually ends with consumption).  Trade off therefore is necessarily intertemporal, the same with leisure.  In fact there seems to be an intuitive link between higher consumption and higher leisure.

Austrian:

The Austrian Business Cycle Theory (ABCT), by contrast, recognizes that monetary injections alter relative prices as they typically enter the capital market first before spreading to the rest of the economy. By lowering the interest rate and bidding up prices for capital goods relative to those of consumer goods, the additional money in the capital market will change the allocation of resources in favor of capital-constructing investments, which will have to be liquidated when the policy of monetary injections is ended.

Friedman --

Workers will initially interpret [an unexpected rise of prices and wages] as a rise in their real wage—because they still anticipate constant prices—and so will be willing to offer more labor (move up [on] their supply curve), i.e. employment grows and unemployment falls. Employers may have the same anticipations as workers about the general price level, but they are more directly concerned about the price of the products they are producing and far better informed about that. They will initially interpret a rise in the demand for and price of their product as a rise in its relative price and as implying a fall in the real wage rate they must pay measured in terms of their own product. They will therefore be willing to hire more labor (move down [on] their demand curve). -- 1976, p.223

In the Friedman-Lucas Model, there will only be prolonged periods of above and below average growth under the unrealistic assumption that people err systematically— continuously overestimating prices in the boom and continuously underestimating them in the bust (Garrison, 1989, p.1 . Under the far more realistic assumption that workers and producers will try to take inflation into account--let alone the new classical assumption of “rational” (i.e. model consistent) expectations--there will be no periods of booms and recessions but only random deviations from trend growth. In sum, fooling people into supplying more of their products and services does not generate cycles.” -- http://www.mises.org/journals/scholar/maanen.pdf

Two major problems with the theory of representative agents:  Even if we eliminate the tails of a distribution capital and labor still do not functions homogenously.  However, through infection across the many vectors of subcategory, the effects are able to offer somewhat general observations (read Freidman’s defense of positive economics). 

?my question: So can’t we understand people to be rational optimally (the survivors) and irrational in the aggregate?  Differentials will change over time.  I don’t know how one would measure this change, but it should theoretically exist.

You cannot underestimate the problem of central bank involvement, if you are going to posit a business cycle.  The central bank acts to make sure that nominal prices (somehow real prices are effected unequally across sectors, or at least until a lag completes its cycle) do not depreciate, as one of the main goals of a central bank.  In another way, if the nominal prices are changed continually as a matter of policy, but the central bank, does this provide a propagation mechanism?

A look specifically at deflation, despite occasional economic sophistry Salerno is one of my favorite Austrians:  http://www.mises.org/journals/scholar/salerno.pdf