Here's what I had to say about it last year, discussing why the U.S. federal government cannot inflate the debt away:
Inflate it away
Hang on, critics say, there is a solution: inflation. If the money depreciates, the debt becomes smaller. For example, if inflation is 5% and real GDP growth is 2%, then the nominal GDP (inflation plus real growth) will be about 7%. That nominal growth will bring in tax revenue, lower the debt-to-GDP ratio, and the situation will quickly go back to what we saw in the 1990s, where rising revenue and slowing spending can quickly stabilize and even reverse the deficit.
It'd be great if things were that simple, but we live in a more complex world. In the first place, as noted above, most of the future deficit is entitlements―spending that adjusts for inflation and will rise sharply due to changing demographics. Parts of the federal budget will rise as fast as the CPI and inflation does nothing to address those costs. Alan Auerbach, an econ professor at the University of California, Berkeley, estimates 90% of the budget would increase along with inflation because future government spending is almost 100% entitlements. If entitlements were reformed by delinking them from inflation, then it would be easier to implement an inflation policy―but no one is talking about ending cost of living adjustments because that would be the end of their political career.
Inflation also requires trapped investors. A 30-year Treasury bond worth $1,000 today at 3% interest will decline in value by about 50% if interest rates climbed to 8%, because the money investors receive back 30-years hence will be greatly devalued. The problem here is that investors (including foreign central banks) don't want to hold 30-year bonds, they are increasingly holding short-term bonds of 2 years or less.
The federal government likes a short-term debt structure because it keeps interest costs low. However, this not only makes inflating near impossible (since the government constantly has to issue new bonds), it is also extremely dangerous. Investors hold short-term debt when they fear that interest rates could rise because they get their money back quickly and can buy new bonds at higher interest rates, avoiding the devaluing scenario outlined above. Even without resorting to an explicit policy of inflation, the federal government is increasingly at risk of a bond market crisis.
This risk is similar to the maturity transformation used by financial companies. There was fear that even blue chip firms such as General Electric (NYSE:GE) could go bust in 2009 because they borrow heavily in the short-term commercial paper market and continually roll it over to finance long-term projects. It is like an old fashioned bank run: your money is lent out to your neighbor for 30-years, but you want your money today. In the case of a financial firm, it borrows 30-day paper at a low interest rate and lends it out at a high rate. If people stop buying that paper, however, the firm is bankrupt.
In other words, even if a firm or country can continue in existence, if it has a lot of short-term debt and for whatever reason, investors today fear it will go bankrupt, then it will go bankrupt because it relies on short-term funding. One of the notable aspects of this type of crisis is the inversion of the yield curve. Short-term interest rates surge, while long-term rates stay lower, because the government faces an immediate financing crisis.
In sum, inflation won't reduce the budget deficit and its effect on the bond market could trigger a crisis.
This isn't news to the Treasury
Despite the government's dismal fiscal picture, the U.S. government is not run by idiots―which is why the U.S. government is considering floating rate notes. From the linked article:
While it may seem like odd timing to start issuing floating-rate debt, since most analysts predict interest rates are unlikely to get much lower, Wall Street analysts say the changes make some sense.
...For one thing, the new debt products wouldn't necessarily increase Treasury's exposure to short-term fluctuations in rates. That is because thefloating-rate notes mainly would be used in place of short-term debt―bills and notes that are issued with maturities of less than two years. Treasury is constantly issuing new bills, mainly to replace maturing debt.
Issuing a floating-rate note of two years in place of a series of three-month bills also would reduce the number of times Treasury would have to sell new debt, which it does via auction. That is appealing, bankers say, especially in light of the recent debt auctions of heavily indebted countries such as Spain and Italy. Those debt sales have been nail-biting events for the financial markets.
In early August 2012, the Treasury announced floating-rate securities are coming, possibly by late 2013.
In adopting this course, the U.S. government tacitly anticipates the exact scenario I mention above. Taking this step is a positive move, since it shows forward thinking, but it won't make inflating the debt away any more realistic. What it does do, however, is potentially avoid a Greek-style panic in the bond market because the government won't have to rollover as much debt.