* Study highlights risk bond buying could lead to losses
* Losses would prevent Fed making payments to Treasury
* Missed payments would risk political interference
By Alister Bull
WASHINGTON, Jan 28 The U.S. Federal Reserve is paying close attention to risks linked to its bond buying program, including the possibility of losses on its massive portfolio that might touch off a political fire storm and harm the central bank's independence.
A Fed staff research paper released last week highlights a range of scenarios under which bond buying in 2013 inflates the danger the Fed incurs a loss.
The central bank regularly returns portfolio profits to the Treasury and it has never missed a payment before. Last year, remittances hit a record $89 billion thanks to income from assets on its balance sheet, which has more than tripled to almost $3 trillion since the financial crisis struck in 2007.
But the researchers warn the Fed could miss payments for up to four years under some scenarios.
The Fed has been buying bonds to drive down longer term borrowing costs after cutting overnight interest rates to near zero in late 2008 to support a fragile economic recovery.
Minutes of the Fed's last meeting in December unexpectedly highlighted growing discomfort with the bond buying, currently running at $85 billion a month, although analysts do not expect any change to be announced after policymakers gather this week.
A Fed loss, while not meaningful in an economic sense, could pose a serious political liability that might expose the central bank to unwelcome scrutiny from U.S. lawmakers, some of whom have been heated critics of the bond buying.
"While this is of little macroeconomic significance, it will not go unnoticed," Charles Plosser, president of the Philadelphia Fed and a noted policy hawk, said in November.
"It is a risk to perceptions about the institution, which eventually may put the Fed's independence at risk."
Other officials have also voiced a keen awareness of the potential costs of the asset purchase program, with Fed Chairman Ben Bernanke going out of his way this month to stress that these costs, while hard to gauge, would not being ignored.
"We have found this to be an effective tool, but we are going to continue to assess how effective because it is possible that, as you move through time and a situation changes, that the impact of these tools could vary," he said on Jan. 14. "When something is more costly, you do a little bit less of it."
The Fed, in a statement scheduled to be released after a two-day meeting at about 2:15 pm (1915 GMT) on Wednesday, is expected to confirm that purchases will be kept up until it sees a significant improvement in the outlook for the labor market.
Minutes of its December meeting showed that several policymakers thought it would be appropriate to slow or stop the purchases well before the end of 2013, while a few thought they would be warranted until the end of the year. A few other policymakers saw a need for considerable policy accommodation, but did not spell out a time frame in which to end the purchases.
The Fed staff research paper examined three scenarios for the balance sheet: no additional increase in its size in 2013, further growth of $500 billion, and further growth of $1 trillion.
The researchers, Seth Carpenter, Jane Ihrig, Elizabeth Klee, Daniel Quinn and Alexander Boote, then calculated what would happen to Fed income as the central bank began to exit from its extraordinary policy measures and raise interest rates.
Assuming rates rose at the pace economists now expect, Fed payments to the Treasury would stay high through 2015, but would then be halted for several years if the balance sheet grew by either $500 billion or $1 trillion in 2013, according to the paper. Under the third scenario, in which no more assets are added in 2013, there was no halt in payments.
If the balance sheet grew by $1 trillion, the research paper calculated that payments would be suspended for four years and the loss would peak at $45 billion in 2019.
The losses would be even deeper if interest rates rose significantly faster than anticipated.
According to the simulation run by the Fed staff, the loss could spike as high as $125 billion in 2019 if 10-year rates were 1 percentage point higher than the consensus of the Blue Chip survey of private economic forecasters, which the paper used as its baseline.
Such losses would not hurt the Fed's ability to conduct monetary policy, the researchers emphasized. Also, they could easily be offset by the benefits of stronger growth if the bond buying succeeded in spurring economic activity, which would broadly lift tax receipts to the Treasury as household income and corporate profits rose.
Furthermore, a halt in Treasury payments would be more than made up by the much larger cumulative Fed remittances to the public purse over the period covered by the study, which goes through 2025.
The paper estimates remittances to total $720 billion from 2009 to 2025, or $40 billion a year, compared with Fed payments to the Treasury of $25 billion a year before the financial crisis. So even if there are losses, the Fed will pay more money to the Treasury over the longer run due to its bond purchases.