Just to add to this discussion, the key aspect about the endogenous nature of money is its ambivalent effects on the working of the economic system. On the one hand, as stressed by many post-Keynesian monetary economists (especially circuitistes and modern monetary theorists), the endogeneity of money enables both the level of investment and growth to surpass what it would otherwise be in a context of self-financing.
According to this view, a recognition of the endogeneity of money frees us from the "fictitious" constraint of a fixed money stock and, as such, opens up new possibilities (from a economic policy standpoint) for achieving full employment and improved living standards (e.g., via public investment financed by government deficit financing and money creation). Also, it forces us to look for a better explanation in regard to the causes of inflation and to reconsider the popular view that inflation occurs solely as the result of an excessive rate of growth in the money supply or as a consequence of government deficit spending. In a context of endogenous money, the causality between increases in prices and the money supply can also be considered as flowing from prices and output to money rather than uniquely the other way around, as is most often believed.
On the other hand, as recently emphasized by the staff economist of the Bank for International Settlements (BIS), the endogenous nature of money, by allowing investment to surpass the capacity of self-financing, also acts to intensify the inherent risks and instability of the modern economy (in which finance plays a critical role) by creating the conditions that lead to unsustainable booms in credit and asset prices that "can eventually lead to serious financial strains and derail the world economy" (Borio and Disyatat, 2011:27).
Now, let me be clear: I'm not saying that these approaches are irreconcilable, or that they exclude each other's views on the issue. On the contrary, one has to look very closely to uncover the difference between the views on the monetary system of post-Keynesian monetary economists and those of BIS economists. They are quite similar in many respects, as recently highlighted by economist Bill Mitchell. For instance, recall that the late Hy Minsky, a post-Keynesian economist, emphasized long ago the destabilizing effect of the modern financial system, a notion that is closely aligned with the views of the BIS economists today. So, in this sense, all I mean to suggest is that the focus of these two groups of economists tends to be different, not that both views are necessarily different in scope.* (For instance, modern monetary economists have been doing some excellent work to address the financial stability issue. See, for instance, Randall Wray and Eric Tymoigne.)
Finally, I will just conclude by saying that, in Canada (where I reside), empirical evidence pointing to the endogeneity of money (i.e., that money supplied by the central bank is demand-led) has been around for a while. Consider this excerpt from Bank of Canada Technical Report 16: Monetary Base and Money Stock in Canada by economists Kevin Clinton and Kevin Lynch arguing against the notion of an exogenous money supply:
...the findings contrary to the monetarist position are strongly enhanced by evidence that emphatically demonstrates causality running from money to the base. The historical association observed between the two arises primarily from the influence of deposits on bank reserves, not vice versa, so that the existing correlation, weak though it may be, could give an exaggerated impression of how well the money supply could be controlled via the base. [...] The empirical tests reject the notion that there is "direct" link between bank reserves and bank deposits and that changes in bank reserves cause changes in bank deposits. (4,40)This technical report was published in 1979. I know of no convincing evidence that refutes these findings (keeping in mind that Canada no longer requires banks to hold reserves).
* The difference between the two approaches lies mainly in their views regarding the existence of the Wicksellian notion of natural rate of interest. Although this is not an insignificant issue, for the purpose of this post there is no need to elaborate further on this point.
Borio, C., and P. Disyatat, Global imbalances and the financial crisis: Link or no link? Bank for International Settlements Working Paper No. 346, May 2011.
Clinton, K. and K. Lynch, Bank of Canada Technical Report 16: Monetary Base and Money Stock in Canada, Bank of Canada, 1979