"Tight money causes unemployment to rise, and loose money causes unemployment to fall. Therefore, if monetary policy had a 20% inflation target rather than a 2% inflation target, monetary policy would be looser, and unemployment would be lower". Wrong, and now obviously wrong. But 40 years ago most economists thought it was right.
"Tight money deflates asset bubbles, and loose money inflates asset bubbles. Therefore, if monetary policy included asset prices in its inflation target, an incipient asset bubble would cause tighter monetary policy, deflating the bubble, and so asset bubbles would be less common". Is that statement equally wrong, and for exactly the same reasons?
"Tight" and "loose" monetary policy need to be defined relative to what is expected, and what people expect will not be invariant to the monetary policy regime.
Let's imagine a conversation that might take place in the future, under a different monetary policy regime, where central banks try to target house prices.
"Should we buy that house? The price does seem rather high, compared to everything else. You don't think it might be a bubble, like in the 2000's?"
"No, it can't be a bubble, because the central bank assured us its new monetary policy would prevent house price bubbles. And in any case, everything else except houses seems to be falling in price, so the only safe investment seems to be in houses. If we don't get into the housing market now, our savings will keep on depreciating, and we never will be able to afford to buy a house."
If the prices of everything else are declining (and importantly are expected to continue to decline) - why wouldn't you just hold cash?
Posted by: reason | July 06, 2009 at 05:25 AM
Nick, reason is right on.
"...everything else except houses seems to be falling in price, so the only safe investment seems to be in houses. If we don't get into the housing market now, our savings will keep on depreciating..."
if there is a general deflation going on then how are their savings deprecitating in real terms?
Posted by: Adam P | July 06, 2009 at 06:55 AM
Yep. I only noticed that after I had posted....
I just forgot to mention: in 2010 they put a tax on cash balances, as the only way to stop hoarding during a deflation and force people to spend, and get us out of the great recession.
Posted by: Nick Rowe | July 06, 2009 at 07:56 AM
Those that have cash are keeping it in cash or cash equivalents. Cash is appreciating relative to falling asset prices (you buy more for less than you would have). One of those assets is your ability to generate income - so as long as your income is higher than your expenses, you are doing great. Unfortunately, many are not in that position, so they experience inflation with respect to their expenses (their income relative to their costs) while their assets continue to deflate.
It is entirely reasonable to have inflation and deflation simultaneously - one money stock is decreasing (circulation) and one is increasing (monetary/banking).
Posted by: pebird | July 07, 2009 at 11:04 AM
I loved your previous post, and will eventually link to it. But this one isn't quite as persuasive. Let's start with a few basic macro principles.
1. The long run path of relative prices should be almost exactly the same under a 4% inflation target as under a 6% inflation target.
2. The long run path of relative prices will differ between a 4% inflation target, a 4% nominal wage rate target, and a 4% NGDP target. Indeed that is precisely why macroeconomists debate the relative merits of these targets.
3. The path of relative prices under a narrow CPI target will differ from the path of relative prices under a broader inflation target that includes asset prices. So your second assertion is not making "exactly" the same error as the first assertion. It may be wrong, but it is not obviously wrong.
One way to understand why is to consider a housing market shock that occurs after wages are set. Under a simple CPI target, the Fed doesn't respond. Under a broader CPI plus target the Fed responds to higher housing prices with a contractionary monetary policy. Because wages are under long term contracts, unemployment rises. This takes the edge off the housing boom.
Posted by: Scott Sumner | July 08, 2009 at 09:06 AM
Scott: I mostly agree. Perhaps I should have re-stated my post:
The second assertion makes an invalid argument. The conclusion may (or may not) be true, but it does not follow from the premise.
Yes, I think it does matter what exact price index or nominal variable the central bank targets. It matters because some prices are more flexible than others, some are more subject to relative shocks than others, and we (both central banks and regular folk) have better information on some prices than on others. But the sort of assertions we hear arguing that house prices (or other asset prices) need to be included in the monetary policy target generally fail to mention such factors. They just assume that agents' decision rules would be the same, while monetary policy changed.
It all comes back to what "loose" and "tight" monetary policy would mean, under a different regime.
Posted by: Nick Rowe | July 08, 2009 at 09:28 AM
Nick, Yes I agree, the basic problem is that a lot of people don't understand the super-neutrality of money, and the distinction between cyclical effects from unanticipated monetary shocks, and secular changes in relative prices which are caused by all sorts of non-monetary factors. So I agree with the thrust of your argument, I just thought the second post on this topic was a bit too categorical in its assertion of money neutrality.
Posted by: Scott Sumner | July 08, 2009 at 09:17 PM