Wow:
$800 billion for perhaps a 1% average decline in the unemployment rate over the next three years.
The long-term cost? Perhaps a persistent 0.1% annual decline in GDP, and certainly a persistent $40 billion in extra interest payments per year.
So even taking CBO baseline as given, I think only hyperbolically discounting voters would support such a plan.
Of course, I believe voters (and certain economists) are hyperbolic discounters…..
But hey, at least we’ll get back to within 1% of potential GDP faster if we pass the plan!
Oh, wait…….
Garett Jones: Your numbers hold if the low estimate is correct.
MarkS
Menzie-
The unemployment rate recovery to 4.8% by 2015 pre-supposes that the baby-boomers leave the workforce as projected by the BLS. I would predict that the erosion of retirement and pension funds will force a large percentage of older workers to remain in the workforce. I believe that the BLS workforce growth projections will have to be updated as the nation adjusts to credit collapse.
I find it most interesting that the CBO anticipates GDP to recover to baseline growth by 2014, three years after most agency, Alt-A and adjustable rate mortgages finish resetting.
Joseph
I like Keynes famous quote “The long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is past the ocean is flat again.
When old sailing ships were caught is a major storm the crew would sometimes throw some of their precious cargo overboard to save the ship. Now, if you were on the upper decks you might complain that in the long run we would all be poorer. But if you were in steerage with water up to your neck you might consider it a worthy trade off.
To the people with jobs, a few million unemployed may seem like no big deal, but to those less fortunate millions it is a life shattering event. You lose your job, you lose your home, and most terrifying of all, you lose your health insurance. One kid with a broken arm, one sick baby, one auto injury and you are wiped out. Some people can ride out the storm in comfort. Others are not so lucky.
Dear Menzie,
Many thanks for your comment and for posting the CBO results. If I did the math right, my claim stands if we average the “low” and “high” estimate to get a central tendency.
I’m guessing the stimulus bill probably can’t generate that big a multiplier in the New Keynesian model CBO is presumably using. Too many transfers and lump-sum tax cuts; too much G in the out-years after prices and wages adjust and crowding out becomes a bigger problem.
If the bill had met Summers’s TTT test (e.g., moved up in time and focusing on G) I’m guessing the CBO numbers would look a bit better…..
Makes me wistful for what might have been: A bill crafted by Summers could’ve been a beauty to behold–too bad Pelosi et al. wrote it instead…..
Garett Jones: I stand corrected. I was looking at 2010 only, at 1.2 ppts. Averaging over three years gives 0.9 ppts.
If I recall aright from my time at CBO, they’re not using a New Keynesian model for their key simulations (although a DSGE might be used as an adjunct), but rather a fairly conventional old style macroeconometric model. In this context, your conjectures are likely correct. However, on the timely/temporary part, the very fact that the output gap is so long and persistent under baseline means that the policy has pretty positive effects.
Another way of putting this is to think about the reduction in the output gap — 2.35 ppts, 2.05 ppts, 0.8 ppts, and 0.35 ppts in 2009, 2010, 2011, 2012 respectively. I think with discounting (plain ol’ exponential even), this looks pretty good — of course conditional on what interest rates will be in the out years.
Apparently the blogging software didn’t like my “less than” symbol so it truncated my message. Here’s a repost without the fancy math notation 🙂
If you average in the high and low then you get a boost in GDP for 2009 of 2.6% and another 2.2% in 2010 for a total of about 4.8%. There are additional benefits in 2011, 2012 and 2013.
If you average in the high and low, it looks like the decline in GDP is about 0.1% in perpetuity after 2015.
In other words, the long run annual cost is less than 1/50th of the short run benefit.
If you discount the future at a rate of interest r less than 2%, then this appears to be a positive NPV investment.
Erik Brynjolfsson: Thanks for the calculation — and welcome onboard! Yes, I think the software takes the less than sign as the beginning of a html tag. Wait until you try to get greek letters in fractions….
Garett Jones
I stand corrected by Brynjolfsson. CBO numbers taken at face value imply that the stimulus bill would pass any conventional cost-benefit analysis. 2% is quite a low real discount rate.
AJB
My background in finance compels me to ask if there are any estimates out there on the NPV of the stimulus package? A more granular real option analysis would be even better.
So much for “Worst Since the Great Depression”, if CBO really projects that lacking the stimulus package unemployment will top at 9%, with an output gap of 7.4%.
In 1982 it was 10.8%, with an output gap of 8.9% (per Gordon). One wonders why there wasn’t such a dramatic rush to a huge fiscal stimulus package then.
Was that a bad mistake?
Markel
In 1982 it was 10.8%, with an output gap of 8.9% (per Gordon). One wonders why there wasn’t such a dramatic rush to a huge fiscal stimulus package then.
Are you under the impression that Volker was running a zero interest rate policy in 1982?
anon
The PV depends on how long the 0.1% drop in growth rates last. If it is a permanent drop in growth rates, then I cannot imagine the PV is positive for any sensible discount rate. I am thinking of the Gordon growth formula 1/(r-g) here and how sensitive it is to g.
I cannot tell how long that drop is suppose to last. If only to 2019, the NPV is big. If forever, then the NPV is surely negative.
But all this is just the formula way of asking the crowding out question…
Mr. Glass surely knows that comparing the current “unemployment rate” to how it was measured during the Great Depression is an apples to oranges comparison. U6 is what really matters in how all of this is playing out. I am, of course, a complete amateur who just follows news and blogs after taking an intro class that combined macro and micro economics many years ago. But am I really wrong on this point?
BillT
Joseph 2-11 8:53 comment…
Right On!!
The figures are meaningless in this arguement if they fail to include the real impact on individuals or collective groups of people effected
Too bad a “Do-Nothing” scenario isn’t included per the Radical-Right’s approach so the CBO
could provide a truely dismal depression forecast
for the coming 8-12 years in my opinion…to compare against.
bt/
Wow:
$800 billion for perhaps a 1% average decline in the unemployment rate over the next three years.
The long-term cost? Perhaps a persistent 0.1% annual decline in GDP, and certainly a persistent $40 billion in extra interest payments per year.
So even taking CBO baseline as given, I think only hyperbolically discounting voters would support such a plan.
Of course, I believe voters (and certain economists) are hyperbolic discounters…..
But hey, at least we’ll get back to within 1% of potential GDP faster if we pass the plan!
Oh, wait…….
Garett Jones: Your numbers hold if the low estimate is correct.
Menzie-
The unemployment rate recovery to 4.8% by 2015 pre-supposes that the baby-boomers leave the workforce as projected by the BLS. I would predict that the erosion of retirement and pension funds will force a large percentage of older workers to remain in the workforce. I believe that the BLS workforce growth projections will have to be updated as the nation adjusts to credit collapse.
I find it most interesting that the CBO anticipates GDP to recover to baseline growth by 2014, three years after most agency, Alt-A and adjustable rate mortgages finish resetting.
I like Keynes famous quote “The long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is past the ocean is flat again.
When old sailing ships were caught is a major storm the crew would sometimes throw some of their precious cargo overboard to save the ship. Now, if you were on the upper decks you might complain that in the long run we would all be poorer. But if you were in steerage with water up to your neck you might consider it a worthy trade off.
To the people with jobs, a few million unemployed may seem like no big deal, but to those less fortunate millions it is a life shattering event. You lose your job, you lose your home, and most terrifying of all, you lose your health insurance. One kid with a broken arm, one sick baby, one auto injury and you are wiped out. Some people can ride out the storm in comfort. Others are not so lucky.
Dear Menzie,
Many thanks for your comment and for posting the CBO results. If I did the math right, my claim stands if we average the “low” and “high” estimate to get a central tendency.
I’m guessing the stimulus bill probably can’t generate that big a multiplier in the New Keynesian model CBO is presumably using. Too many transfers and lump-sum tax cuts; too much G in the out-years after prices and wages adjust and crowding out becomes a bigger problem.
If the bill had met Summers’s TTT test (e.g., moved up in time and focusing on G) I’m guessing the CBO numbers would look a bit better…..
Makes me wistful for what might have been: A bill crafted by Summers could’ve been a beauty to behold–too bad Pelosi et al. wrote it instead…..
Garett Jones: I stand corrected. I was looking at 2010 only, at 1.2 ppts. Averaging over three years gives 0.9 ppts.
If I recall aright from my time at CBO, they’re not using a New Keynesian model for their key simulations (although a DSGE might be used as an adjunct), but rather a fairly conventional old style macroeconometric model. In this context, your conjectures are likely correct. However, on the timely/temporary part, the very fact that the output gap is so long and persistent under baseline means that the policy has pretty positive effects.
Another way of putting this is to think about the reduction in the output gap — 2.35 ppts, 2.05 ppts, 0.8 ppts, and 0.35 ppts in 2009, 2010, 2011, 2012 respectively. I think with discounting (plain ol’ exponential even), this looks pretty good — of course conditional on what interest rates will be in the out years.
Would the average of the high and low be a mean in the bell curve sense? And the high/low would be.. what? 3 standard deviations from the mean?
Apparently the blogging software didn’t like my “less than” symbol so it truncated my message. Here’s a repost without the fancy math notation 🙂
If you average in the high and low then you get a boost in GDP for 2009 of 2.6% and another 2.2% in 2010 for a total of about 4.8%. There are additional benefits in 2011, 2012 and 2013.
If you average in the high and low, it looks like the decline in GDP is about 0.1% in perpetuity after 2015.
In other words, the long run annual cost is less than 1/50th of the short run benefit.
If you discount the future at a rate of interest r less than 2%, then this appears to be a positive NPV investment.
Erik Brynjolfsson: Thanks for the calculation — and welcome onboard! Yes, I think the software takes the less than sign as the beginning of a html tag. Wait until you try to get greek letters in fractions….
I stand corrected by Brynjolfsson. CBO numbers taken at face value imply that the stimulus bill would pass any conventional cost-benefit analysis. 2% is quite a low real discount rate.
My background in finance compels me to ask if there are any estimates out there on the NPV of the stimulus package? A more granular real option analysis would be even better.
So much for “Worst Since the Great Depression”, if CBO really projects that lacking the stimulus package unemployment will top at 9%, with an output gap of 7.4%.
In 1982 it was 10.8%, with an output gap of 8.9% (per Gordon). One wonders why there wasn’t such a dramatic rush to a huge fiscal stimulus package then.
Was that a bad mistake?
In 1982 it was 10.8%, with an output gap of 8.9% (per Gordon). One wonders why there wasn’t such a dramatic rush to a huge fiscal stimulus package then.
Are you under the impression that Volker was running a zero interest rate policy in 1982?
The PV depends on how long the 0.1% drop in growth rates last. If it is a permanent drop in growth rates, then I cannot imagine the PV is positive for any sensible discount rate. I am thinking of the Gordon growth formula 1/(r-g) here and how sensitive it is to g.
I cannot tell how long that drop is suppose to last. If only to 2019, the NPV is big. If forever, then the NPV is surely negative.
But all this is just the formula way of asking the crowding out question…
Mr. Glass surely knows that comparing the current “unemployment rate” to how it was measured during the Great Depression is an apples to oranges comparison. U6 is what really matters in how all of this is playing out. I am, of course, a complete amateur who just follows news and blogs after taking an intro class that combined macro and micro economics many years ago. But am I really wrong on this point?
Joseph 2-11 8:53 comment…
Right On!!
The figures are meaningless in this arguement if they fail to include the real impact on individuals or collective groups of people effected
Too bad a “Do-Nothing” scenario isn’t included per the Radical-Right’s approach so the CBO
could provide a truely dismal depression forecast
for the coming 8-12 years in my opinion…to compare against.
bt/