4 ms·

The charts you link make the argument that the output gap was huge. Notably, 2021Q4 CBO output gap is -0.44, about where it was in 2004Q4, when we were already

The charts you link make the argument that the output gap was huge. Notably, 2021Q4 CBO output gap is -0.44, about where it was in 2004Q4, when we were already well on to recovery. You can plug the actual value into the Taylor rule and it comes out at about 12.5%. My takeaways are that: a.) Interest rates are going way up from here. b.) Even if the Fed moves to where the dot-plots are projecting - about 3% in 2024 - it is way too little to contain inflation. I think their assumption is that inflation will come down on its own as supply-chain issues abate, but I don't share that assumption with the Fed.

JumpCrisscross 5y ago
> You can plug the actual value into the Taylor rule and it comes out at about 12.5%. OP's equation for the Taylor Rule is...unorthodox [1]. For those who don't want to calculate equilibrium real interest rates and take logarithms of linear trends of GDPs, the Fed has a heat map tool [2]. It says 7 to 9%, using defaults. Higher than 3%. But within the model's error bars. (All that said, I have more confidence in inflation being manageable than rates remaining below 3%. Trading on that hunch, however, would require fighting the Fed. You don't fight the Fed.) [1] https://en.wikipedia.org/wiki/Taylor_rule#As_an_equation https://en.wikipedia.org/wiki/Taylor_rule#As_an_equation [2] https://www.atlantafed.org/cqer/research/taylor-rule https://www.atlantafed.org/cqer/research/taylor-rule
RC_ITR 5y ago
Have you considered that we are in the midst of an unprecedented 'goods and services" supply shock and that will drive inflation 100x more than interest rates (which haven't shown any correlation to inflation since the 80's)?
- memling 5y ago
  > (which haven't shown any correlation to inflation since the 80's) Side question: how do we calculate that correlation? To clarify, I understand that you can take two sets of observations and calculate rho, etc. What I'm really curious about is how (or whether) the correlation is adjusted. Suppose there is always some kind of base inflation that doesn't correlate with interest rates; do we have a sense for what the additive effect of changing interest rates does?
  - RC_ITR 5y ago
    https://fred.stlouisfed.org/graph/fredgraph.png?g=NhJO https://fred.stlouisfed.org/graph/fredgraph.png?g=NhJO Sorry to clarify - you can run any correlation on those two lines (wither lagged or not) and you'll find very little direct relationship, but the more important point is: By saying interest rates and inflation are not correlated, I meant that both inflation and interest rates were declining to historic lows, which is in direct opposition to the theory that 'raising interest rates reduced inflation.' As you can see on this chart, interest rates aren't even that much lower than they were in the period where we were worried we'd never see inflation ever again. My main gripe with the current narrative around inflation is 'does it really seem logical that moving from 2.5% to 1.0% 10Y yields is what caused inflation (when going from 7.5% to 2.% had no effect), given all we know about how broken our supply chains have become?'
    - hcknwscommenter 5y ago
      It was not just the 10Y though. QE + traditional federal reserve cuts affected the whole curve, out to 30 and even longer-dated (non-USG) instruments. Then you had the giant stimulus on top of that. This was about as close to helicopter money as I think the US has ever done. Many other countries adopted similar measures. We are seeing the effects of inflation from each of the above. And now, the Fed is winding down QE and raising interest rates. Stimulus is fading. This all takes time. But yes, I think it is pretty reasonable to assume inflation will go down over the next 12 months. Edit to change QT to QE
      - RC_ITR 5y ago
        1) 10Y by far is the most important benchmark used in the US economy. It's the foundational rate for almost all consumer lending including mortgages and widely used as the RFR in discounting cashflows. 2) My gripe with the stimulus argument is 'why is inflation accelerating 12 months after we stopped'? For example, are used cars (a major contributor to current inflation) really going up in price b/c people can afford to pay higher prices, or is it because we are making way fewer new cars? Same story for houses (given that we've chosen to abandon billions of sqft of office space and replace that with at-home offices). Sure, loose money support isn't helping inflation, but the idea that the solution to our current inflation problem is to decrease demand seems unthoughtful, given the fact that we know we are producing well below our potential (on a global scale). I actually don't disagree with you that inflation will go down in the next 12 months, but I think it will primarily be on the supply side. At least I hope so.
- truffdog 5y ago
  Something else that stopped in the 1980s was the oil crisis. Possibly we all give Volcker too much credit.

JumpCrisscross 5y ago

> You can plug the actual value into the Taylor rule and it comes out at about 12.5%. OP's equation for the Taylor Rule is...unorthodox [1]. For those who don't want to calculate equilibrium real interest rates and take logarithms of linear trends of GDPs, the Fed has a heat map tool [2]. It says 7 to 9%, using defaults. Higher than 3%. But within the model's error bars. (All that said, I have more confidence in inflation being manageable than rates remaining below 3%. Trading on that hunch, however, would require fighting the Fed. You don't fight the Fed.) [1] https://en.wikipedia.org/wiki/Taylor_rule#As_an_equation https://en.wikipedia.org/wiki/Taylor_rule#As_an_equation [2] https://www.atlantafed.org/cqer/research/taylor-rule https://www.atlantafed.org/cqer/research/taylor-rule

RC_ITR 5y ago

Have you considered that we are in the midst of an unprecedented 'goods and services" supply shock and that will drive inflation 100x more than interest rates (which haven't shown any correlation to inflation since the 80's)?

memling 5y ago
> (which haven't shown any correlation to inflation since the 80's) Side question: how do we calculate that correlation? To clarify, I understand that you can take two sets of observations and calculate rho, etc. What I'm really curious about is how (or whether) the correlation is adjusted. Suppose there is always some kind of base inflation that doesn't correlate with interest rates; do we have a sense for what the additive effect of changing interest rates does?
- RC_ITR 5y ago
  https://fred.stlouisfed.org/graph/fredgraph.png?g=NhJO https://fred.stlouisfed.org/graph/fredgraph.png?g=NhJO Sorry to clarify - you can run any correlation on those two lines (wither lagged or not) and you'll find very little direct relationship, but the more important point is: By saying interest rates and inflation are not correlated, I meant that both inflation and interest rates were declining to historic lows, which is in direct opposition to the theory that 'raising interest rates reduced inflation.' As you can see on this chart, interest rates aren't even that much lower than they were in the period where we were worried we'd never see inflation ever again. My main gripe with the current narrative around inflation is 'does it really seem logical that moving from 2.5% to 1.0% 10Y yields is what caused inflation (when going from 7.5% to 2.% had no effect), given all we know about how broken our supply chains have become?'
  - hcknwscommenter 5y ago
    It was not just the 10Y though. QE + traditional federal reserve cuts affected the whole curve, out to 30 and even longer-dated (non-USG) instruments. Then you had the giant stimulus on top of that. This was about as close to helicopter money as I think the US has ever done. Many other countries adopted similar measures. We are seeing the effects of inflation from each of the above. And now, the Fed is winding down QE and raising interest rates. Stimulus is fading. This all takes time. But yes, I think it is pretty reasonable to assume inflation will go down over the next 12 months. Edit to change QT to QE
    - RC_ITR 5y ago
      1) 10Y by far is the most important benchmark used in the US economy. It's the foundational rate for almost all consumer lending including mortgages and widely used as the RFR in discounting cashflows. 2) My gripe with the stimulus argument is 'why is inflation accelerating 12 months after we stopped'? For example, are used cars (a major contributor to current inflation) really going up in price b/c people can afford to pay higher prices, or is it because we are making way fewer new cars? Same story for houses (given that we've chosen to abandon billions of sqft of office space and replace that with at-home offices). Sure, loose money support isn't helping inflation, but the idea that the solution to our current inflation problem is to decrease demand seems unthoughtful, given the fact that we know we are producing well below our potential (on a global scale). I actually don't disagree with you that inflation will go down in the next 12 months, but I think it will primarily be on the supply side. At least I hope so.
truffdog 5y ago
Something else that stopped in the 1980s was the oil crisis. Possibly we all give Volcker too much credit.