4 ms·
> This Cantillon effects [...] is not accurate in modern financial markets > The FED intervenes by buying financial assets which directly benefits people who h
by lazulicurio 6y ago
> This Cantillon effects [...] is not accurate in modern financial markets
> The FED intervenes by buying financial assets which directly benefits people who hold these financial assets [...] Poor people without savings (and without financial assets) don't benefit
Would you mind expounding on this a little bit? From my perspective, you seem to be decrying the Cantillon effect in one sentence yet acknowledging its existence in the next.
Edit: To explain my confusion a bit more: In my (albeit limited) understanding, the Cantillon effect claims that monetary injections will spread unevenly across the economy. So, I read your comment as: "The claim that monetary injections will spread unevenly across the economy is not accurate. The reality is much simpler. The FED just injects money in a way that it spreads unevenly across the economy."
- jcfrei 6y agoI think the explanation the Cantillon effect gives is wrong. The monetary injection spreads uneven not because the effects on purchasing power are delayed between different groups but because it mostly increases the value of financial assets. And rich people have a larger share of their wealth in financial assets. Most of the wealth in the lower income percentiles is probably in consumer goods (appliances, cars, etc) and very little in savings / financial assets.
- lazulicurio 6y agoI'm still having trouble grasping the distinction that you're trying to make. Is it just that there's a separation between purchasing power and asset value?