Bob Murphy wasn't happy with this post. These sorts of discussions, like arguments over Herbert Hoover, can really reveal what different universes people are from.
He presents this graph against my claim that "We haven’t seen any fiscal stimulus since 2009 (and that was underwhelming)":
It's clear what Bob is seeing here. He's seeing an increase in government spending. What I'm seeing is during the biggest macroeconomic event since the 1930s, government spending declines, rises quickly back to trend, and then continues on trend through several quarters of high unemployment (the growth after mid-2009 even looks a little slower than trend to me, but let's just call it "back to trend"). That sounds like the definition of "no fiscal stimulus since 2009" to me, and yet this is supposed to be some massive refutation. If your policy during the boom years is comparable to your policy during the bust years it seems best to just say "you aren't doing any macroeconomic policy with your fiscal decisions" - which is precisely what people like me have been complaining about.
The real expenditure figures drive the point home:
This is from the BEA's obnoxious new website. The nominal figures seem to match up with Bob's FRED figures - I imagine the FRED data is from the BEA. [UPDATE - Unfortunately I closed some other excel files I was working with, but take a look at the expenditure data back to 1995 - the slowdown relative to earlier decades is very clear].And as a contrast, this is FRED data on both nominal and real government spending - with real spending going back to the 1930s:
See that thing that happened in the 1940s? That is fiscal stimulus. I am not telling Bob he has to be happy about what we're doing right now. I know he's not. Returning to trend isn't his thing - he wants to reverse the trend. I know that. But that doesn't magically make this a fiscal stimulus, and I'm going to continue to be frustrated with people who say "we've tried that and it doesn't work". We haven't tried it. We had a little burst in early 2009 that got us back to trend, and then we've been about on trend ever since. What we've tried is not fiscal stimulus. If you want to say "we've tried continuing on trend and that doesn't seem to help", then by all means say that (the empirical economist in me would say that's still wrong - counterfactuals and all that - but it's better than what people usually say).
*****
In a criticism of this post, Jonathan Catalan basically makes a bunch of points I agree with and talk about a lot on here. He writes:
"In the case of Obama, it is not about jobs created versus jobs saved. It is about opportunity cost. And, neither is it about whether or not Obama can create a short-run positive increase in employment. For the sake of simplicity, less us assume that private investment is zero for a period of six months. If government redistributes this capital from entrepreneurs to the public sector and then invests it in some form of public work there will likely be net job creation during that period of time. But, if we were to judge this in terms of benefit to the economy as a whole we would be mistaken to correlate job creation with economic growth.
The point I try to make in my article “Government Spending is Bad Economics” (that government cannot economize) is lost here. It is not about job creation. It is whether or not the job creation that comes from government spending, if any, is worth the loss which is represented by what could have been had that capital not be redistributed towards less preferred ends."
Of course it's about opportunity costs - to put it another way, "It is whether or not the job creation that comes from government spending, if any, is worth the loss which is represented by what could have been had that capital not be redistributed towards less preferred ends." This is the point I make repeatedly when I review empirical studies of fiscal stimulus. The only thing that Jonathan and I differ on is what we think is the likely answer to the question.
In a criticism of this post, Jonathan Catalan basically makes a bunch of points I agree with and talk about a lot on here. He writes:
"In the case of Obama, it is not about jobs created versus jobs saved. It is about opportunity cost. And, neither is it about whether or not Obama can create a short-run positive increase in employment. For the sake of simplicity, less us assume that private investment is zero for a period of six months. If government redistributes this capital from entrepreneurs to the public sector and then invests it in some form of public work there will likely be net job creation during that period of time. But, if we were to judge this in terms of benefit to the economy as a whole we would be mistaken to correlate job creation with economic growth.
The point I try to make in my article “Government Spending is Bad Economics” (that government cannot economize) is lost here. It is not about job creation. It is whether or not the job creation that comes from government spending, if any, is worth the loss which is represented by what could have been had that capital not be redistributed towards less preferred ends."
Of course it's about opportunity costs - to put it another way, "It is whether or not the job creation that comes from government spending, if any, is worth the loss which is represented by what could have been had that capital not be redistributed towards less preferred ends." This is the point I make repeatedly when I review empirical studies of fiscal stimulus. The only thing that Jonathan and I differ on is what we think is the likely answer to the question.
Posts
You might also want a look at this graph:
ReplyDeletehttp://2.bp.blogspot.com/-wGidGoBkZs8/Tb62FWP4IbI/AAAAAAAAAf8/f50QkdfzJ18/s1600/stimulus1.PNG
You know, Cahal - I seem to remember a certain Nobel laureate pointing out in 2009 that the structure of the stimulus was such that it would be a net drag on GDP starting in early 2011... :)
ReplyDeleteJust so we're on the same page, do you realize that if you believe real spending has been flat, then you take the growth in nominal spending as equal to inflation, meaning we've seen about a ~12% price increase over the worst economic crisis since the GD?
ReplyDeleteI guess that means you're no longer in a different world than the one I inhabit on my trips to the grocery store...
Given the overall largeness of the federal government these days I'm not even quite sure what "stimulus" would mean. From its inception DHS, etc. has spent roughly $1 trillion on homeland security measures. That's $200 billion more more than the so-called stimulus. That $1 trillion has gone in part into the construction or renovation of all sorts of buildings and facilities. So it seems like we've had a lot of stimulus (even if it wasn't viewed that way) for some time now.
ReplyDeleteA trillion over ten years in good economic times for the most part spent by consolidated agencies that existed before the DHS and would have spent money on stuff anyway doesn't seem like it's all that relevant to the issue at hand.
ReplyDeleteThe DHS came into being in Nov. of 2002; OHS was spending money before that. And no, w/o the creation of the DHS you would have likely seen a rather different amount of spending (indeed, probably less spending). By consolidation DHS became like the DoD, a 800 lb. gorilla. Given our experience with consolidated vs. non-consolidated agencies it becomes clear that the former are better when it comes to bureaucratic knife fighting and at squeezing funds from the Congress. Given the theory of iron triangles, etc. etc. etc., that ought not be surprising.
ReplyDeleteAnyway, the point is that the government already spends scads of money; any "stimulus" is more than likely going to get lost in that noise (which doesn't even get into issues like timeliness, etc.). You are more than welcome to tell me why that it isn't the case.
Gary, of course they spend more money then they otherwise would have spent. I never said they wouldn't have.
ReplyDeleteThe point is a lot would have been spent already - the net increase was a lot less than the 1 trillion, and it was spread out over ten years mostly during a good economy. I don't know what you're talking about when you refer to this as a stimulus.
Daniel,
ReplyDeleteNo, what you wrote was this: "...would have spent money on stuff anyway..." Meaning with or without the creation of the DHS; and my response is, well, no. They would not have spent that money (the full $1 trillion) anyway without the creation of the DHS. The creation of the DHS extended the sort of possible spending horizon for things associated with security at home.
"The point is a lot would have been spent already - the net increase was a lot less than the 1 trillion, and it was spread out over ten years mostly during a good economy."
This ignores the context of that timeline though. Was the economy "good" because it was "artificially inflated" in part because of all the nat. sec and defense spending (as well as by other measures undertaken by the government)?
"I don't know what you're talking about when you refer to this as a stimulus."
So during "good times" spending by the government does not stimulate the economy, but during "bad times" it does?
re: "Meaning with or without the creation of the DHS; and my response is, well, no"
ReplyDeleteGary I'm not in the mood to play this game. They would not have "spent that money" (your words). They would have "spent money" (my words). If your "gotcha" has me looking like a complete idiot there's probably something wrong with your "gotcha". I get things wrong, but I'm not a complete idiot - it's all there in plain English.
re: "So during "good times" spending by the government does not stimulate the economy, but during "bad times" it does?"
"Good times" is obviously vague, but as a rule of thumb we expect government spending - which is not allocated efficiently because it does not use the price mechanism - to crowd out private spending. So whatever the government spends in "good" times ought to be well justified by things like externalities and things with strong welfare-enhancing qualities.
Indeed, go with this line of reasoning a bit further, it seems clear to me that one could at least hypothesize that an economy can only take so much "stimulus" before you start to see a diminishing return or even resistance to it. I'd say that is in part because the more the government spends the more it is likely to horn in on areas where it will anger people. And this isn't so much an issue of "crowding out" spending by private entities as simply getting involved in areas where people are not expecting the state to be.
ReplyDeleteRight... people don't like the government being involved in places where they don't like the government being involved.
ReplyDeleteThis seems obvious... I'm waiting for the other shoe to drop. What is the point exactly?
Daniel,
ReplyDeleteNo gotcha, just not a terribly useful way to describe the actual process that goes on. Honestly, there is no reason to take offense, so I'm not quite sure why you are.
That just begs the question though. Was what was spent during the "good times" well justified during the Bush administration?
Daniel,
ReplyDeleteYou are right, it is obvious. But whether it seems obvious to you or me is not the point I'd say; whether D.C. is tone deaf on the matter is the issue (and D.C. is often quite tone deaf - in significant part merely due to inertia).
Dude - you were trying to make it look like I said they would have spent it all anyway. I'm not taking offense - I'm just pointing out you do this sort of thing of trying to put words in my mouth all the time.
ReplyDeletere: "Was what was spent during the "good times" well justified during the Bush administration?"
There's clearly a ton of god-awful unjustified stuff spent in every administration since the beginning of the republic. All I can say is that it's better than the alternatives. It's better than a dictator empowered to do stuff that we don't have a say in, and it's better than being told that we the people have to live in a society where we can't democratically provide for ourselves. It doesn't always work out well - but it's better than the alternatives. To prevent things getting spent with poor justification, we write constitutions to constrain government, and we check and decentralize power. That solution works for me, until we come up with something better.
On the gotcha thing -
ReplyDeleteDo you notice a pattern of you accusing me of saying things that I repeatedly have to show you I've never said? That's what I'm referring to.
I know I frustrate you too. But often what frustrates you about me is interpretative. I have analyzed libertarianism - for example - in a way that frustrates you. I understand that that does. Those sorts of conflicts happen, and I think I've got a good reason for making claims I do so we just have to work through that.
But to have to deal with being told that I've made an assertion that the record obviously shows I've never made day after day is a "gotcha" commenting M.O.
"Dude - you were trying to make it look like I said they would have spent it all anyway."
ReplyDeleteWell, you could just say, no, that is not what I meant, and be done with it (that's what I do when you misinterpret me). The other option is to argue over the matter and get all huffy. And you are totally jumping to conclusions about the "gotcha" stuff. So I'll pretend that the former is the approach you took and say, well, ok, now we're on the same page.
"There's clearly a ton of god-awful unjustified stuff spent in every administration since the beginning of the republic."
Well, to me that is exactly why Keynesianism fails. A political system does not yet exist for it to work properly in other words, and may never exist.
"Do you notice a pattern of you accusing me of saying things that I repeatedly have to show you I've never said?"
ReplyDeleteI'm not "accusing" you of anything. If I ever accuse you of something I'll say it like this: J'accuse!
Anyway...
Daniel,
ReplyDeleteI'm a bit confused about the talk of fiscal stimulus 'cause the definition seem to shift a lot (not saying your definition is shifting though). Sometimes it's increase in budget deficit, increase in government spending, increase in government spending relative to GDP and so on. Your definition seem to be increase in spending relative to trend.
Is there a standard way to identify fiscal stimulus or is it every economist for himself?
re: "Well, you could just say, no, that is not what I meant, and be done with it"
ReplyDeleteHaha - OK, I just have to point out that that's what I tried to do at 10:33 PM and you responded with a five line paragraph!
re: "Well, to me that is exactly why Keynesianism fails. A political system does not yet exist for it to work properly in other words, and may never exist."
I don't follow the logic. Exactly how are you defining Keynesianism here?
Per -
ReplyDeleteWell many actions can be stimulative, right? Those include increasing spending and increasing the deficit. Increasing spending without increasing the deficit means revenues are going up, so you are taking money from people that would have demanded a certain portion of it - so while the spending might be stimulative it certainly would be less stimulative.
My point with looking at things relative to trend is just to look at the magnitude of the response. Government spending grows over time, just like consumer spending and investment spending. If government spending grows as much as it would have without a recession, it doesn't seem right to me to suggest that we are adding to demand with government spending, does it?
Fiscal stimulus is an increase in government spending. It is more stimulative if it is more reliant on deficits. So the question is "increase relative to what"? A reasonable counterfactual seems to me to be "what we would have spent anyway if nothing went wrong". Certainly you don't want to point to business as usual and call it not business as usual simply because business as usual increases government spending at a low, steady rate.
What's wrong with saying "fiscal stimulus is counter-cyclical"? Isn't that the case? That the government should spend as much/more even when GDP and revenue falls?
ReplyDeleteHas spending since 2007 been counter cyclical?
Daniel,
ReplyDeleteI beg to differ, but as long as we're on the same page, it doesn't matter.
I'm defining it as spend like a drunken sailor in bad times and reign in spending like a miser in good times.
That a government could run up so much debt so quickly and apparently not do any significant fiscal stimulus is an amazing dysfunctional political system.
ReplyDeleteIf Keynesian stimulus works, at least the Lord Skidelsky flavor, should we not have seen a response in economic performance from the uptick in spending during early 2009?
ReplyDeleteI also agree with Silas above: spending growth may 'only' be back to trend, but that still gives us significant spending increases.