It seems that rising housing prices increases credit expansion due to higher nominal values of collaterals. The chain from money to housing prices seems less clear. Here, a little bit of theory apart of measuring is of help: Once upon a time there was an enduring debate between so-called bullionists and anti-bullionists, arguing about what was later called the real bills fallacy. According to the real-bills-doctrine, credit expansion based on sound and short-termed collaterals just accompanied the "needs of trade" and ipso facto could not be inflationary. What the anti-bullionists did not understand, however, is the fact that the nominal prices of collaterals depend on the money supply. As T.M. Humphrey from the FED in Richmond writes:
"[The Real Bills Doctrine] links the nominal money stock to the nominal volume of bills, a variable that moves in step with prices and so the money stock itself. By linking the variables, it renders both indeterminate. Far from preventing overissue, it ensures that any random jump in prices will, by raising the nominal value of goods-in-process and so the nominal volume of bills presented as collateral for loans, cause further increases in borrowing, lending, the money stock, spending, and prices ad infinitum in a self-perpetuating inflationary spiral. In short, the doctrine fails to perceive that price increases themselves expand the needs of trade and so generate—and justify—the very monetary expansion necessary to perpetuate them. The doctrine’s flaw consists of the dynamically unstable price-money-price feedback loop established when money is allowed to be governed by the needs of trade. Far from prohibiting monetary inflation, the real bills mechanism virtually guarantees it."Thus, the answer the the question posed by Greiber and Setzer is: both, the increase in money supply causes the asset and housing bubble (because a general, economywide bubble can only be induced by money growth, since otherwise increase in demand can only mean increase in relative demand), which in turn causes more money growth due to lower risk premia and credit expansion.
It is clear, however, that the ultimate cause must be on the money supply side, because this is the only way we can shift upwards aggregate nominal demand in a fractional reserve system in which actual hoarding is quite low. But to understand this, monetarists have to go beyond Friedman's so-called 'restatement' of the quantity theory as a theory of money demand and remember the genuine approach to the quantity theory as a theory of the price level, equalizing real aggregate demand and real aggregate supply. To do so, they have to give up the Phillips-curve (no matter if short- or long-run) as an alternative explanation of the price level. In doing so, the can avoid all the cost-push arguments they usually employ - a line of argument fiercefully attacked by genuine neoclassical proponents of the quantity theory like Wicksell and Fisher.
