I'm curious as to whether or not Japan's EFFR graph looks similar. What I wonder about is the influence do to the fact that US T-Bonds are in competition with other sovereign bonds, or even commodities, or other investment vehicles. That seems to be a point where the Fed's will might have a limit. Risk-free is a huge positive, but when there are far greater returns to be had, the demand will have to fall.
Regarding your first question, here's a chart that should answer it. The "LIBOR based on Japanese Yen" field is pretty much equivalent to whatever the Bank of Japan might publish. (I don't actually know the BoJ's name for its overnight interbank lending rate.) There are subtle differences---here's a comparison of the effective fed funds rate to the US dollar LIBOR---but it's mostly about methodological minutiae. Regarding the latter question... Commercial banks are not allowed to engage in investment (or speculative) behavior with their reserve (consumer-facing) accounts, so I believe they're restricted to holding cash, treasuries, loans, mortgage-backed securities, and I think some corporate bonds/paper. But, either way, with reserve accounts, banks must hold low-risk assets. I believe banks are also not supposed to hold assets denominated in foreign currencies, though I'm less sure there. Edit: For clarification, since those FRED graphs can be a bit hard to interpret, in the first graph, blue is the Japanese T-bill rate, and red is the Japanese yen LIBOR. I wish the Bank of Japan had a nice data plotting tool like the St. Louis Fed, so I'd have a bit more confidence in the correctness of those plots...
