4 ms·

The official was correct, from the perspective of economy-wide price stability. The problem wasn't inflation, the problem was expensive food. Obviously, this p

The official was correct, from the perspective of economy-wide price stability. The problem wasn't inflation, the problem was expensive food. Obviously, this particular exchange was of no comfort to those who suffer from high food prices, but why should monetary policy be the remedy for high food prices (to the exclusion of its current goals) instead of looking for remedies in fiscal policy, trade policy, regulatory policy, or other instruments better suited to a more targeted redress?

SilasX 6y ago
Well, there's technically correct, and substantively correct. Most people would take away the lesson that "hey, the current definition of inflation doesn't quite capture the reality of normal people's cost of living", not "computer says no suffering, sorry bro". But I don't think it's even technically correct from the perspective of price stability. The inflation the central bank cares about is the kind that is the result of (to simplify) "too much money chasing too few goods". Technological deflation is totally orthogonal to that, and should be ignored to the extent possible. Furthermore, even on its on terms, the calculation isn't correct. A fifty percent faster CPU doesn't mean the effective price has fallen fifty percent, because the CPU is only occasionally the bottleneck. (And do they ever adjust inflation upward for quality reductions?)
- fennecfoxen 6y ago
  The BLS does not include a CPU-megahertz (or gigahertz) in its market basket. It uses CPU speed (and other characteristics) to divide the market into "high-end", "mainstream", or "economy" computers, then substitutes new items in the correct slot when as the new items become available, rather than waiting for the price of the older computer to plummet.
  - SilasX 6y ago
    That's still equating technological deflation with (absence of) money-induced inflation, and it still assumes that the more powerful computer can displace the effects of food price inflation (or that the out-of-date computer can keep up with the same present needs). Also, Dudley said: >>“Today you can buy an iPad 2 that costs the same as an iPad 1 that is twice as powerful,”
    - fennecfoxen 6y ago
      > it still assumes that the more powerful computer can displace the effects of food price inflation Well, yes-kind-also-no? It's not an index that describes the cost of necessities or human survival. It's an index that describes how consumers spend money on goods and services, such that the value of the money remains more or less constant. In that sense, yes, people's computer-money spending patterns can affect the price of food. Literally anything else in the index can do so as well. No, the index is not "assuming" anything can "displace" food; it's just never set out to be a thing where food is this sort of baseline. I can't be 100% sure, but you seem to think that there should be an index which cares more about food. Moreover, you seem to think that more public policy should revolve around this price index rather than the vanilla CPI that is currently used. However, I would suggest that monetary policy is far blunt an instrument. It is one of the least-targeted tools possible, with the largest set of possible side effects, and food is too small a slice of the economy. There's a reason we go for economy-level price stability instead of sector-specific. Imagine if the US had a real bumper crop, more corn and soybeans than you could eat: to meet food price stability targets the Fed would be obligated to shrink the money supply until the corn was as expensive as it used to be, and everyone's mortgages, loans, and credit cards would be harder to pay off.
      - SilasX 6y ago
        I think I addressed these points in the other subthread and agreed that, yes, in general, technological gains can indeed alleviate budget problems, including affording food, but it doesn't hold for the example he gave. The comment for a longer reply. https://news.ycombinator.com/item?id=26390606 https://news.ycombinator.com/item?id=26390606 >There's a reason we go for economy-level price stability instead of sector-specific. Imagine if the US had a real bumper crop, more corn and soybeans than you could eat: to meet food price stability targets the Fed would be obligated to shrink the money supply until the corn was as expensive as it used to be, and everyone's mortgages, loans, and credit cards would be harder to pay off. That I don't have a problem with. The concern is with the opposite scenario: say the Fed is pumping enough money to raise food prices 20%, but techno wizards increase production efficiency across the board by 20% and exactly cancels it out. In that scenario the Fed should say, "wow, we're pumping too much money", not "inflation is just right", because they've stolen credit for the gains of technology.
anm89 6y ago
You are playing a word game here by taking a very specific definition of inflation useful to bureaucrats to argue that ipad processor speeds meaningfully counteract food prices in the way an average person perceives the price level. This simply isn't true unless again you redefine all of those words, which, while not hard to do, isn't particularly useful to people who aren't that specific group of bureaucrats. So pick whatever definition of inflation you want. Just understand that what normal people care about in terms of inflation is how many pieces of paper they need to achieve the basic goals of their life and your definition is totally irrelevant to their set of concerns. Your having a totally separate conversation then they are with nothing but some semantic overlap.

SilasX 6y ago

Well, there's technically correct, and substantively correct. Most people would take away the lesson that "hey, the current definition of inflation doesn't quite capture the reality of normal people's cost of living", not "computer says no suffering, sorry bro". But I don't think it's even technically correct from the perspective of price stability. The inflation the central bank cares about is the kind that is the result of (to simplify) "too much money chasing too few goods". Technological deflation is totally orthogonal to that, and should be ignored to the extent possible. Furthermore, even on its on terms, the calculation isn't correct. A fifty percent faster CPU doesn't mean the effective price has fallen fifty percent, because the CPU is only occasionally the bottleneck. (And do they ever adjust inflation upward for quality reductions?)

fennecfoxen 6y ago
The BLS does not include a CPU-megahertz (or gigahertz) in its market basket. It uses CPU speed (and other characteristics) to divide the market into "high-end", "mainstream", or "economy" computers, then substitutes new items in the correct slot when as the new items become available, rather than waiting for the price of the older computer to plummet.
- SilasX 6y ago
  That's still equating technological deflation with (absence of) money-induced inflation, and it still assumes that the more powerful computer can displace the effects of food price inflation (or that the out-of-date computer can keep up with the same present needs). Also, Dudley said: >>“Today you can buy an iPad 2 that costs the same as an iPad 1 that is twice as powerful,”
  - fennecfoxen 6y ago
    > it still assumes that the more powerful computer can displace the effects of food price inflation Well, yes-kind-also-no? It's not an index that describes the cost of necessities or human survival. It's an index that describes how consumers spend money on goods and services, such that the value of the money remains more or less constant. In that sense, yes, people's computer-money spending patterns can affect the price of food. Literally anything else in the index can do so as well. No, the index is not "assuming" anything can "displace" food; it's just never set out to be a thing where food is this sort of baseline. I can't be 100% sure, but you seem to think that there should be an index which cares more about food. Moreover, you seem to think that more public policy should revolve around this price index rather than the vanilla CPI that is currently used. However, I would suggest that monetary policy is far blunt an instrument. It is one of the least-targeted tools possible, with the largest set of possible side effects, and food is too small a slice of the economy. There's a reason we go for economy-level price stability instead of sector-specific. Imagine if the US had a real bumper crop, more corn and soybeans than you could eat: to meet food price stability targets the Fed would be obligated to shrink the money supply until the corn was as expensive as it used to be, and everyone's mortgages, loans, and credit cards would be harder to pay off.
    - SilasX 6y ago
      I think I addressed these points in the other subthread and agreed that, yes, in general, technological gains can indeed alleviate budget problems, including affording food, but it doesn't hold for the example he gave. The comment for a longer reply. https://news.ycombinator.com/item?id=26390606 https://news.ycombinator.com/item?id=26390606 >There's a reason we go for economy-level price stability instead of sector-specific. Imagine if the US had a real bumper crop, more corn and soybeans than you could eat: to meet food price stability targets the Fed would be obligated to shrink the money supply until the corn was as expensive as it used to be, and everyone's mortgages, loans, and credit cards would be harder to pay off. That I don't have a problem with. The concern is with the opposite scenario: say the Fed is pumping enough money to raise food prices 20%, but techno wizards increase production efficiency across the board by 20% and exactly cancels it out. In that scenario the Fed should say, "wow, we're pumping too much money", not "inflation is just right", because they've stolen credit for the gains of technology.

anm89 6y ago

You are playing a word game here by taking a very specific definition of inflation useful to bureaucrats to argue that ipad processor speeds meaningfully counteract food prices in the way an average person perceives the price level. This simply isn't true unless again you redefine all of those words, which, while not hard to do, isn't particularly useful to people who aren't that specific group of bureaucrats. So pick whatever definition of inflation you want. Just understand that what normal people care about in terms of inflation is how many pieces of paper they need to achieve the basic goals of their life and your definition is totally irrelevant to their set of concerns. Your having a totally separate conversation then they are with nothing but some semantic overlap.