We hear a lot about the need for macroeconomics to be consistent with microeconomic foundations, that's been a driving force behind macroeconomic modeling for the past several decades. But those foundations are not fully adequate for the problems that macroeconomists face.
Microeconomists generally assume that prices move to clear markets. Their main interest is in understanding how equilibrium outcomes change when there are changes in the economic environment. Importantly, microeconomists almost always assume that prices move immediately to clear all markets, and this is where the relationship and usefulness of microeconomic foundations for macroeconomic theory is called into question.
Exactly how prices do this isn't well specified. Instead microeconomists make use of an abstraction -- the Walrasian auctioneer -- who calls out prices, assesses the outcome in terms of excess demand or excess supply in all markets, readjusts prices, computes a new assessment of excess supply and demand, and this continues until demand equals supply in all markets. This is assumed to happen immediately, i.e. prices are assumed to move frictionlessly and timelessly to clear all markets.
But these models don't help a lot when, as assumed in modern macroeconomic models, the problem is that prices are sticky and the economy is in transition from one long-run equilibrium to another. In these models, agents are assumed to do the best that they can given the sticky, sub-optimal prices. Thus, macroeconomists follow microeconomists and assume that all markets clear in the short-run (an assumption that can be questioned, especially at a time like now). But prices still differ from their optimal long-run values. The sticky price assumption prevents prices from moving to clear all markets at their long-run equilibrium values, so the economy is not in a long-run equilibrium. Even if the auctioneer knows what prices ought to prevail, some prices are stuck in the short-run making the optimum unattainable.
Macroeconomists need to know more about the process that moves the economy from one long-run equilibrium toward another as economic conditions change (e.g. due to a shock to mortgage markets). This is where microeconomists can provide more help. The key thing that we need -- how markets operate when the Walrasian auctioneer cannot impose the equilibrium price vector immediately, but instead must change prices slowly over time -- is not well understood. How do markets operate and interact under these conditions?
One of the lessons from macroeconomics is that prices do not always move fast enough to keep markets in balance, and this observation -- this macroeconomic foundation -- needs to guide microeconomic theorists. The Walrasian auctioneer is a convenient theoretical assumption, somehow prices always magically move immediately to clear all markets and attain the highest possible societal welfare (and it's useful for comparing long-run outcomes). But that assumption is not very useful as a foundation for macroeconomic models that try to explain short-run swings in the macroeconomy. Microeconomic and macroeconomic models ought to be consistent with each other, but the microeconomic foundations need to have realistic assumptions about the ability of price changes to clear markets.