Empirical Tests of Monetary Policy Measurements and Their Implications for Macroeconomics

(ProQuest: ... denotes formulae omitted.)INTRODUCTIONNonborrowed reserves have started to become negative since 1959 (figure 1), which was the first time in the U.S. history.The formula the Federal Reserve used to calculate nonborrowed reserves up to December 12, 2007 was:Nonborrowed reserves = Total reserves - Discount Window BorrowingsThen on December 12, 2007, the Federal Reserve started using the following formula:Nonborrowed reserves = Total reserves - Discount Window Borrowings- TAF borrowingThe negative value of nonborrowed reserves was because total borrowings were larger than the total reserves. Not all TAF borrowing was included in total reserves, thus negative nonborrowed reserves occurred. I have sent e-mails to the Federal Reserve asking how much TAF borrowing was hold in total reserves. Unfortunately, the staff at the Federal Reserve could not give me a satisfactory answer.If the Federal Reserve changed the accounting method for nonborrowed reserves intentionally, then did they also change the monetary policy indicator? Although the Federal Reserve claimed that the federal funds rate is used as the monetary policy instrument, the Federal Reserve does not always do as it says. Thus it makes sense to recheck the monetary policy indicator.How does the policy stance today compare with what it was in earlier periods? Furthermore, this paper will use the corrected nonborrowed reserves and re-evaluate various implications of the monetary policy. Last but not least, this paper will be an additional support for Barnett's proposal (2012) for creating the Fed's own data bureau by examining another Federal Reserve's inaccurate data.As a matter of fact, nonborrowed reserves are not the only inaccurate data that is inaccurate.INACCURATE DATA IN THE FEDAbnormal nonborrowed reservesSuddenly on January, 1st, 2008, the nonborrowed reserves of U.S. banks became negative. Then they increased to as large as 486 billion in June, 2009 (figure 1).First of all, this was the first time nonborrowed reserves had been in a negative number. How can borrowed reserves exceed total reserves? It is an accounting error. A simple example: If there is a 6 inch apple pie on the table, what the Federal Reserve was doing was taking an 8 inch apple pie from the original one.Secondly, the large increase in nonborrowed reserves later was due to the fact that total reserve was expanded by the Federal Reserve, but the Fed was still using an inaccurate accounting method which failed to include all the Term Auction Facility (TAF) borrowing in the total reserves.Poor monetary-aggregate dataThe Federal Reserve still uses the simple sum monetary aggregate data rather than Divisia index monetary aggregate data which has already been applied by many other countries. Simple sum aggregate data is inaccurate. You cannot compare apples to oranges. The Federal Reserve cannot combine money in the checking account to savings account, since they have different costs known as the user cost (Barnett & Serletis, 2000). It costs more to hold the money in a checking account than in a savings account. The bank has to be compensated for providing extra liquidity if one holds money in a checking account.No pre-sw eeps dataBarnett (2010) pointed out that Ml aggregates are far below actual data. Banks only provide the Federal Reserve post-sweeps checking account data but no pre-sweeps data. In order to provide the Federal Reserve with less required reserves, banks usually transfer checking account deposits into savings account. In this case, the Federal Reserve is not able to monitor the exact liquidity, since money in the checking accounts is one of the most important channels to provide liquidity. To be accurate, they should have both pre-sweeps and post-sweeps data.The Fed and the financial crisisBoth Barnett and Chauvet (2011) and Taylor (2008) showed that the monetary excesses were the main cause of the recent financial crisis. …