Vimothy had a great comment here which directly relates to the importance of the accounting.
“ I came across a paper in this vein while trawling through Google Scholar recently. The paper is Cebula (1995), “The impact of federal government budget deficits on economic growth in the US”. I haven’t had chance to read it properly yet, but from the conclusion:
The primary conclusions regarding the impact of fiscal policy variables on per capita real economic growth are:
a) federal government purchases appear to exercise a negligible impact on economic growth;
b) the federal budget deficit acts to significantly reduce the economic growth rate;
c) a higher maximum level of federal government personal income tax rates significantly reduces the economic growth rate; and
d) a higher maximum level of federal corporate income tax rates significantly reduces the economic growth rate.
Richard Cebula probably did some serious and diligent research. He probably used some very credible and reasonable math to come to the conclusion he did.
But we can’t trust the conclusion at all. Here is an excerpt from the Godley paper:
“There is an obvious shortcoming to the original Christ formula in that it only applies to a closed economy. This deficit is easily remedied by adding exports ot the governmetn expenditure (injections) and imports to taxes (leakages). So the open-economy version of the Christ concept of fiscal stance is given by government outlays plus exports divided by the average tax rate plus the open-economy counterpary of the tax rate, namely , the share of income that leaks abroad. We call this ratio the augmented fiscal stance” (AFS), which now includes the effect of net export demand with the government’s net injections.”
Godley then constructs a time series which looks very much like NGDP out of the AFS. Of course, it’s slightly off. So he does a bit of reasearch of why it is off, and it turns out the errors are nearly exactly mirrored by the level of private (non-financial) lending.
In short, Godley demolishes Cebula, and has a more coherent and simpler explanation on how and why he is correct.
This isn’t a mystery. It’s just a matter of using the accounting and sector balances correctly. Cebula probably did a good job, he’s just not aware of how the entire economy fits together. His conclusions are wrong because he didn’t follow the accounting which matters to GDP.
Does government spending help or hurt the economy? Well, it depends. But you can’t tell from looking at government spending vs. economic activity as a single variable regression. It’s like trying to measure the amount of water in a leaky bucket under an open water spigot.
I don’t have all the links but the IMF has doing a mea-cupla – subtly.
Check the first link in Krugman’s post http://krugman.blogs.nytimes.com/2012/10/09/deleveraging-shocks-and-the-multiplier-sort-of-wonkish/
Haven’t seen the Cebula paper but here is why such analyses go wrong.
The budget deficit is a poor measure of the stance of fiscal policy. The deficit could expand simply because the economy is going into a recession and without the government actually changing its *policy*.
Imagine if the US government were to relax fiscal policy by say increasing expenditures and the private sector were to react to this by increasing its expenditures.
This means that total taxes (as opposed to the tax rate) will be higher and the budget deficit may actually start decreasing.
If one were to just see deficit versus gdp or growth, one would wrongly conclude that lower deficit has led to the growth!!
When you walk through the logic of the Godley paper, it’s a bit like getting smacked on the head.
1. Hey, for GDP to grow, something has to go into deficit!
2. Let’s take into account everything which goes into determining GDP!
3. Oh Yeah, there should be a lag!
4. Wow! it works great!
This is classic too:
“The budget deficit is a poor measure of the stance of fiscal policy. The deficit could expand simply because the economy is going into a recession and without the government actually changing its *policy*.
”
It is easy to lose sight of the fact the U.S. has designed fiscal policy to be countercylical in the structure of how it budgets money. This doesn’t even have to be on the spending side- as taxe receipts ebb and flow with business cycles, deficits will go up during times of low economic growth.
What is “the Godley paper” you mention? What title?
From beowulf:
Fiscal Policy Will Matter (which Godley wrote in 1998 with George McCarthy). I can’t find a free link for it but here’s the JSTOR link.
http://www.jstor.org/discover/10.2307/40721807?uid=3739616&uid=2&uid=4&uid=3739256&sid=21101296466237
In other news: economists noted that there is very strong correlation between doctor visits and people being sick. It is recommended to stop doctors from visiting to improve the health of the population.
In other news economists observe a correlation between gdp and the sale of shoes. To improve gdp, they recommend more production of shoes. In fact they want NGDP targeting. So they recommend controlling shoe production at 4% increase per year.
They will do this by flooding the market with leather, rubber soles, and laces.
If this doesn’t work, Ben “The Cobbler” Bernanke will begin buying shoes until the economy improves.
Shoes are neutral in the long run
It’s also important that the CME begin trading in shoe futures. We need a reliable way to guage market expectations of shoe production.
It’s a 100,000 pair contract, standard size 11, black leather shoe. They can be delivered to zappos, amazon, DSW, or The foot Locker or Payless.
The Shoe Illusion.
http://recom.org/remagv2/wp-content/uploads/2011/07/film_salamanda_july.jpg
“This isn’t a mystery. It’s just a matter of using the accounting and sector balances correctly.”
What if CA deficit = gov’t deficit plus private deficit is incomplete?
And, “1. Hey, for GDP to grow, something has to go into deficit!”
I think which entity dissaves and how matters.
It seems to me that with the sectoral balances there is a stock of medium of exchange that flows. If more medium of exchange is required, does it matter how it is created?
It seems like it does matter how the money is created. My first blush take on it would be there are times when private debt is better, times when government debt is better, and times when the foreign sector needs to go into deficit.
Do I have a hard and fast rule on this? No. But the only way to even begin answering this question is through the lens provided by Godley.
What if the solution to too much debt is not more private debt or more gov’t debt? What if the foreign sector does not want to cooperate (no other country wants to decrease net exports)?
There are some problems which do not have easy answers. I don’t want to start a pissing match, but I’d say this question is part of a series of questions which lead us to start MR.
Look at Japan – what is their solution? They have plenty of government debt, they have decent private debt. What do they do?
Clearly, Japan has a breakdown between the money creation process and real investment.
If you recall, MR had several posts on the S = I + (S-I) relationship as being extremely important.
No pissing match just that if the “solutions” don’t work then maybe the problem is not defined correctly. From what I’ve read by Richard Koo, Japan has a similar problem but not the same. I’m going to focus on the USA. I haven’t read all the S = I + (S-I) posts.
Here is what I think part of the problem is.
http://www.federalreserve.gov/boarddocs/speeches/2002/20021121/default.htm
“The sources of deflation are not a mystery. Deflation is in almost all cases a side effect of a collapse of aggregate demand–a drop in spending so severe that producers must cut prices on an ongoing basis in order to find buyers.1”
“1. Conceivably, deflation could also be caused by a sudden, large expansion in aggregate supply arising, for example, from rapid gains in productivity and broadly declining costs. I don’t know of any unambiguous example of a supply-side deflation, although China in recent years is a possible case. Note that a supply-side deflation would be associated with an economic boom rather than a recession.”
I’m pretty sure all those deflations are price deflations. I also believe bernanke is assuming the definition of economics is true, unlimited wants/needs and limited resources. I don’t believe the definition of economics. If bernanke believes the definition of economics, he thinks whatever real AS is that is what real GDP should be. If real AD is below real AS, it is an aggregate demand shock, and aggregate demand shocks are dealt with by creating more debt. From what I can tell, bernanke has no model for real AS being above real AD without it being an aggregate demand shock.
I also believe bernanke does not have a model where a supply-side deflation would be associated with an economic boom and then an economic bust because of how new medium of exchange is created.
Nice one, Mike.
My thoughts are roughly as follows:
1, Why should deficits be essential for growth? That doesn’t make sense to me.
2, If deficits really are essential for growth, that suggests that a strong relationship between the two should be fairly easy to see in the data. Is it? And if not, why not?
Now, Cebula looks at some data and finds a negative relationship. Maybe you don’t like his method or his sample. That’s okay. But it’s not enough to say that you don’t like it. You need to say why. If Godley is right, why isn’t the relationship positive?
You really need to read the Godley paper if you have not already. He answers these questions in the paper.
Do you have a copy? Ramanan had it up for a few days. Let me know.
1) medium of exchange
2) if not, private debt or current account
“2, If deficits really are essential for growth, that suggests that a strong relationship between the two should be fairly easy to see in the data. Is it? And if not, why not?”
Ragu Rajah pointed out, “starting in the early 1970s, advanced economies found it increasingly difficult to grow. Countries like the US and the United Kingdom eventually responded by deregulating their economies… More routine, once well-paying, jobs done by the unskilled or the moderately educated were automated or outsourced.”
This can explained, I believe, by Bill Vickrey’s point that the US economy had 2 x GDP in available private investments but 3 x GDP in available private savings.
“If governments fail to fill the gap and meet the demand for assets by issuing an adequate volume of securities, the attempt by individuals to acquire assets by non-spending will cause a reduction in sales, temporary investment in excess inventories, cutbacks in orders, unemployment, and reduced national income and product. This may be partially offset by the bidding up of asset values, leading to a certain amount of additional spending out of capital gains, but the “saving” imbedded in these capital gains does not involve the creation of new capital or the employment of individuals in construction.”
The post-war economy was so strong because the investment demand gap was met by the 120% or so of GDP in T-bonds issued during World War II, but as the debt was paid down (and perhaps because the public investments it bought wore out), the economy became sluggish.
http://monetaryrealism.com/why-are-bond-yields-in-a-30-year-bear-market/#comment-8015
Mike,
My memory of Godley’s paper (which I scanned) is that he not actually claiming that deficits are essential for growth, but that for the economy to grow, the government’s fiscal stance has to rise.
That also doesn’t make sense to me. When I think about what determines growth, I think of things like the accumulation of human capital, not a mechanical relationship that has the government simply leaning on the lever of its expenditure relative its average tax rate. Instead, I would have thought that government expenditure is naturally going to rise as the country becomes more wealthy, and people demand the better quality public goods that they are now able to afford–i.e., the causality goes in the opposite direction.
With that said, Godley’s series does seem to track GDP pretty well, so what’s going on there? I haven’t really got the time to go digging into the paper, but when I looked at the graph, what struck me was that the fiscal stance doesn’t just track GDP, it more or less is GDP, with some very small fluctuations about the trend. There’s almost no variation.
To me that suggests: multicolinearity. We saw that with a balanced budget, the fiscal stance and GDP are one and the same thing (because if G = T, then G/t = G/(T/Y) = Y, and so the multicolinearity is perfect–you’re just regressing y on y). If we add in government borrowing, then we might expect to see some cyclical fluctuations around the long run trend as the budget goes in and out of balance. Which we in fact do see. So, I guess it doesn’t seem like a very big deal to me.
Well, my reading of the paper was the accounting identities add up to GDP. This of course is always true.
Then, because we know they add up, we can make projections for the future based on the component parts for GDP. We can find ways to model future levels of fiscal stance, future levels of private lending, future levels of exports and imports.
Then we have useful predictions for what GDP might be.
But you have another issues- why are deficits necessary? NOMINAL deficits are necessary if we want NOMINAL GDP growth. Read the quote from the paper in the first Godley post again, because this is where he states this.
If all the deficit flow variables go to zero, we end up with a steady state stock of NGDP. It’s a matter of the accounting.
Is this case desirable? I do not think rampant price deflation is a good way to promote real investment, because of the riskless opportunity for increasing individual wealth by offered by sitting on your hands.
At the end of the paper, they talk about debt deflation being the only scenario which makes sense, given the state of affairs in 1998.
Of course, Greenspan put off the day of reckoning by pushing rates extremely low and blowing up the credit bubble to many times its size in 1998.
How does an economy have debt deflation is there is no debt?
EDIT TO: if there is no debt?
“NOMINAL deficits are necessary if we want NOMINAL GDP growth.”
The 2 quarter lag was an average but it did vary. I wonder if the lag period for any given quarter correlated with inflation rate (If we were running a Volcker-style 10% inflation rate, this quarter’s nominal deficit would be reduced in real terms by 5% after 2 quarters).
For some reason this all reminds me of Flow5 (who should write a paper called Fiscal Policy Will Not Matter).
“Contrary to economic theory, & Nobel laureate, Dr. Milton Friedman, monetary lags are not “long and variable”. The lags for monetary flows (MVt), i.e. the proxies for (1) real-growth, and for (2) inflation indices, are historically, always, fixed in length. However, the FED’s target, nominal gdp?, varies widely…
since the lags for (1) monetary flows (MVt), & , are synchronous & indistinguishable, by using simple algebra, economic prognostications are infallible (for less than one year). “
So can you translate this into English for me, Beo :
(It actually looks illogical to me, but of course it can’t be)
“What about time lags? The lag between “the AFS at one time and the flow of GDP at another must be governed, under purely hydraulic principles, by the size of the net stock of assets generated by total inflows (government outlays plus exports) less outflows (taxes plus imports). Just as the length of the pro- duction period (approximately one quarter of a year in the United States) may be estimated from the ratio of inventories to production, so the mean lag of GDP behind the AFS will be given by the normal ratio (if there is such a thing) between the stock of net financial assets (government debt plus net overseas assets) and total inflows, that is, government expenditure plus exports.
A “normal” stock-flow ratio is not directly observable. It is, however, the case that the net stock of these financial assets, measured ex-post facto, has hovered around 50 percent of the annual flow of government outlays plus exports for the past twenty years. So, for this period at least, the nominal GDP should track the AFS one for one, with a mean lag of only half a year. This is the amount of time that must elapse before the inflows become outflows, and it is well within the time span that would make the AFS relevant for policy-making purposes.”
Also did not understand this.
I must, in good conscience, outsource my job to India. Ramanan has so much deeper an understanding of Godley’s work than I do (as seen downthread) that he’s the one who should be unpacking this.
Okay I will post two pages of scans from Godley’s book and an article of his where he shows this. (Sometime when I get some time to scan)
Michael, what I am struggling with is why we can extrapolate from the steady-state case. Why does that disprove vimothy’s theory about accum of human capital etc driving increases in the fiscal stance etc.
This is not an extrapolation from a steady state case. It’s a think-through of the implications for stocks when flows are zero. If the flows go to zero, the stocks can’t grow.
Also, this has very little to do with the reasons the fiscal stance might rise or fall. It’s primarily about getting the setup correct so we can begin to investigate the world in a useful manner.
“Michael, what I am struggling with is why we can extrapolate from the steady-state case. Why does that disprove vimothy’s theory about accum of human capital etc driving increases in the fiscal stance etc.”
The steady state case is just a quick way of getting what a good definition of fiscal stance is.
It is true that supply side factors improve the potential output but they may not actually improve the actual output. This is because economies are demand constrained. Consumers do not care if human capital suddenly improved or not. If they have lower incomes, they will not increase consumption.
Firms won’t be able to sell more because they are constrained by how much they call sell. An individual firm can increase its market share but by reducing the market share of another if demand doesn’t keep up.
If the government actually doesn’t exploit the supply side factors, I am not sure what increases in human capital can bring. It is not exploited. Imagine if there is some major advancement in technology. If the government cuts its expenditure and/or increases tax rates there won’t be any advantage because of the advancement. Output will actually fall.
It was Keynes’ observation that economies are demand constrained. (Actually Kalecki but Keynes had the personality to convince). Vimothy’s story is the usual neoclassical supply sider story “supply creates its own demand”. Economies are always at their potential best and fiscal policy can only do bad.
The Keynesian revolution completely changed the profession and actually the way the world is run. After world war 2, economies were run under Keynesian principles – till it fell apart because of the Monetarist counterrevolution.
“It is true that supply side factors improve the potential output but they may not actually improve the actual output.”
Explanations:
1) Entities don’t want what output is being produced
2) Entities don’t want to buy any more output
3) Wealth/income inequality ; some entities have enough output & other entities don’t have enough medium of exchange to purchase the output
“This is because economies are demand constrained.”
It seems to me economies can be demand constrained or supply constrained. It depends.
What if the accounting is missing an entity?
Sorry why multicollinearity?
G is exogenous and t is also exogenous. These are set by the government.
Also G is not equal to T.
The intuition for this has been derived from the fact that in the simplest models, in the “long run” the government budget is balanced, Y is G/t. So G/t has something to do with output roughly.
Hi Ramanan,
G and t are both set by the government, but they’re only really exogenous in the context of simple models. In the “true” model–i.e., reality–they’re not exogenous at all. It would have been impossible, for example, for the US government of 1912 to have the same level of expenditure as the US government of 2012.
G will be equal to T when the budget is balanced. This refers to an example in the paper that someone (Beowolf?) quoted in the comments. If you notice, my statement was conditional: if the budget is balanced, then the G/t = Y. Well, if G/t = Y, there’s no point regressing one on the other: they’re the same series, so of course they’re correlated. Equally, you might note that GDP is highly correlated with C + I + G. But so what? There’s no information about causality to be gotten from such things.
Vimothy,
True. But even if the government cannot obviously quintuple its expenditure overnight, it has a lot of room to set total expenditures.
“G will be equal to T when the budget is balanced. ”
The definition of the fiscal stance G/t does not rely on whether G is equal to T or not. It was originally derived by Carl Christ that in the long run the budget is balanced and G = tY and hence Y = G/t. That was the original intuition.
“then the G/t = Y. Well, if G/t = Y, there’s no point regressing one on the other: they’re the same series”
They are NOT the same series as the graph shows.
G /t is the government expenditure by tax rate. Y is the output. These are separate things. However it happens that these are really close to each other (to be more precise one has to consider the external trade).
Again G is NOT EQUAL to T.
“Equally, you might note that GDP is highly correlated with C + I + G. But so what? There’s no information about causality to be gotten from such things.”
Yeah, but a return so what?
There is a clear relationship between how the government expenditure has steered the economy.
There is a basic confusion here. The expression G/t was derived originally by Carl Christ and hence the reference to the way it was derived.
G is NOT EQUAL to T in real life.
G/t can in theory be anything. For example government expenditure = $1T and tax rate = 20% and consequently the budget in surplus because Y happens to be say $15T.
However this is not the case. It is fiscal policy which drives the economy. If fiscal policy were to not drive the economy G/t (or the more general one including trade performance) and Y would be totally different.
They are not very different implying fiscal policy actually drives the economy.
Once again Again G is NOT EQUAL to T!
To be more precise start from scratch.
Plot G/t (where G is the government expenditure and t the tax RATE) and Y.
G is not equal to tY and hence G/t and Y are totally two different things.
Now see the plot – They are close to each other!!!
Ramanan,
True. But even if the government cannot obviously quintuple its expenditure overnight, it has a lot of room to set total expenditures.
So in practice, whatever freedom the government has is terms of setting its level of expenditure is determined by things like the productive capacity of the economy, the economy’s ability to restructure that productive capacity at will, the effect of taxation on production and so on.
The definition of the fiscal stance G/t does not rely on whether G is equal to T or not. It was originally derived by Carl Christ that in the long run the budget is balanced and G = tY and hence Y = G/t. That was the original intuition.
Right. IF the budget is balanced THEN the fiscal stance equals output. So what we have is a series that IS output, give or take the budget balance. Why should we care about this, given that it doesn’t contain any information? Take my analogy with the GDP identity. We know that GDP = C + I + G — plus or minus some other stuff in an open economy. So we can say with great confidence that C + I + G is going to track GDP, and vice versa. But that’s not very interesting, for the same reason that it tracks it GDP: GDP is just a linear sum of C, I and G.
Now, the worst possible thing to do here would be to reason from the ex post identity that links output to the components of aggregate expenditure and say, “since output is equal to C, I and G, to increase output, just increase G.” I hope we can all agree that that would be very bad economics.
They are NOT the same series as the graph shows.
I didn’t say that they were the same series. I said that IF the budget is balanced, THEN they are the same series. They are almost the same series.
G /t is the government expenditure by tax rate. Y is the output. These are separate things.
They’re different letters, but they’re not really separate things, as you can see from the graph. Y is G plus some other stuff, and t is the government’s share of Y. Say that the government gets a constant share of Y. Over time, Y is growing. If G grows at the same rate as Y and given that t is constant, that’s obviously a stable outcome in some sense. (You could say the same for private expenditure and after tax income, of course). But the idea that this proves that the growth in the fiscal stance “causes” output to rise doesn’t make a lot of sense to me. For one thing, increases in the fiscal stance caused by increasing G can just as easily be recreated by holding G constant and shrinking t. Eventually G will become arbitrarily small relative to the economy–no more government, another “stable” outcome.
Yeah, but a return so what?
So you can’t use something like this to test whether there is a causal relationship between the growth in expenditure and growth in output. They’re necessarily highly correlated.
There is a clear relationship between how the government expenditure has steered the economy.
Well, that’s your hypothesis.
I think that private expenditure drives output growth. To see this, note the correlation between (C + I)/(1 – t) and Y. For the economy to be in a stationary equilibrium, the private sector’s income and expenditure must grow at the same rate. Therefore, private expenditure growth drives output growth, QED.
Vimothy,
“So in practice, whatever freedom the government has is terms of setting its level of expenditure is determined by things like the productive capacity of the economy, the economy’s ability to restructure that productive capacity at will, the effect of taxation on production and so on.”
True. It is true that ultimately fiscal policy has a constraint brought about by the capacity to produce, but neoclassicals and Neokeynsians all argue as if the economy is running at its maximum productive capacity and that fiscal policy has little room. It is really a dangerous ideology – totally damaging.
Since economies are demand-constrained, the actual output need not be equal to the potential output. A boost is needed to bring it close to it.
In fact more production increases supply side factors and there is a positive feedback effect due to this.
“We know that GDP = C + I + G — plus or minus some other stuff in an open economy. So we can say with great confidence that C + I + G is going to track GDP, and vice versa. But that’s not very interesting, for the same reason that it tracks it GDP: GDP is just a linear sum of C, I and G.”
Look (G + X)/(t+mu) is not equal to GDP. So your analogy is not right
“They’re different letters, but they’re not really separate things, as you can see from the graph. Y is G plus some other stuff, and t is the government’s share of Y. Say that the government gets a constant share of Y. Over time, Y is growing. If G grows at the same rate as Y and given that t is constant, that’s obviously a stable outcome in some sense. (You could say the same for private expenditure and after tax income, of course). But the idea that this proves that the growth in the fiscal stance “causes” output to rise doesn’t make a lot of sense to me. For one thing, increases in the fiscal stance caused by increasing G can just as easily be recreated by holding G constant and shrinking t. Eventually G will become arbitrarily small relative to the economy–no more government, another “stable” outcome.”
Oh shrinking t ! Chicago school! Stable outcome?
Unfortunately that misses distribution issues. Decreasing tax rates is positive for GDP but it is a trickle-down economy approach. The government is not just in the business of changing expenditure and tax rates. It is also responsible for how its finances are run, where the expenditures need to be directed in etc. Government investment is an important example.
This kind of argument is for the reduction of the role of the government in the economy.
“So you can’t use something like this to test whether there is a causal relationship between the growth in expenditure and growth in output. They’re necessarily highly correlated.”
The government expenditure is under the control of the government and so is the tax rate. I don’t know how you suppose that the causality is the other way round.
Does activity in the economy talk to computers in the government department to change its expenditure or tax rates?
“I think that private expenditure drives output growth. To see this, note the correlation between (C + I)/(1 – t) and Y. For the economy to be in a stationary equilibrium, the private sector’s income and expenditure must grow at the same rate. Therefore, private expenditure growth drives output growth, QED.”
Consumption is not exogenous. It is dependent on income. Investment has some autonomous element to it but clearly its depends on animal spirits of entrepreneurs and the kind of uncertainty they may perceive in the future.
What the Keynesian revolution brought about was that even if I is exogenous, it is not necessary that it will bring in full employment. Why would entrepreneurs invest if their sales are not improving?
What is worse was that when the article was written, economists were arguing for tightening fiscal policy – as if it doesn’t matter.
“Well, that’s your hypothesis.”
Clear in the chart.
Ramanan,
neoclassicals and Neokeynsians all argue as if the economy is running at its maximum productive capacity and that fiscal policy has little room.
Neoclassicals might think that fiscal policy isn’t a useful stabilisation tool, for various reasons. But I don’t think that they argue that the economy is always at full capacity. In fact, I know they don’t.
Since economies are demand-constrained, the actual output need not be equal to the potential output. A boost is needed to bring it close to it.
Well, it all depends, doesn’t it? If the economy is operating below potential because of a demand shock and the government can usefully do something, then it should. But all economies aren’t always in demand-led recessions.
But this is a separate issue. We were discussing the determinants of long-run growth, not the proper role of the government during the business cycle.
Oh shrinking t ! Chicago school! Stable outcome?
It’s not supposed to be a serious argument, just to show that using the exact same reasoning as you I can derive the opposite result.
The government expenditure is under the control of the government and so is the tax rate. I don’t know how you suppose that the causality is the other way round.
Other way round than what?
Think of it like this: t tells you the government’s income share. If the government’s income share is constant then its tax revenue will grow at the same rate as GDP. If government expenditure grows at a different rate, then this process will become unsustainable in the limit: either government borrowing will explode or government will effectively be destroying output.
Consumption is not exogenous. It is dependent on income. Investment has some autonomous element to it but clearly its depends on animal spirits of entrepreneurs and the kind of uncertainty they may perceive in the future.
Don’t follow your reasoning here. Consumption depends on income to an extent, but it obviously has some kind of “autonomous” component, because people gotta eat, and we know from experience that consumption is the most stable component of expenditure over the cycle.
But clearly, consumption depends on income, just like government expenditure. If government expenditure / private expenditure grows much faster than its take in taxes / disposable income, the process is not sustainable over long horizons.
“But clearly, consumption depends on income, just like government expenditure. If government expenditure / private expenditure grows much faster than its take in taxes / disposable income, the process is not sustainable over long horizons.”
The government has the powers to make its expenditure independent of income. In recent times, the US government’s tax revenues fell. Did it automatically induce a reduction in government expenditure?
“Don’t follow your reasoning here. Consumption depends on income to an extent, but it obviously has some kind of “autonomous” component, because people gotta eat, and we know from experience that consumption is the most stable component of expenditure over the cycle.”
Stable?
Yes there is an kind of autonomous components but this is trivial in comparison to how much people spend. Are you making the case that the autonomous part is much more important than the induced part. i.e., are you arguing that income really does not matter?
“Think of it like this: t tells you the government’s income share. If the government’s income share is constant then its tax revenue will grow at the same rate as GDP. If government expenditure grows at a different rate, then this process will become unsustainable in the limit: either government borrowing will explode or government will effectively be destroying output.”
Nope.
Haven’t you studied the multiplier effect?
In fact the expression G/t is just an improvement of the standard textbook Keynesian multiplier.
An increase in government expenditure will increase output and due to the multiplier effect, bring in higher tax revenues because of higher economic activity.
There are other ways in which the government expenditure can bring about an exploding debt dynamics, not in the way you suggest.
Also, Tax revenues grow at the same rate as GDP in the long run.
The immediate effect will be to improve the tax inflow at a faster rate than the growth rate.
“Neoclassicals might think that fiscal policy isn’t a useful stabilisation tool, for various reasons. But I don’t think that they argue that the economy is always at full capacity. In fact, I know they don’t.”
Not sure! Recently they may have changed. But the atmosphere around the Clinton era was that the surplus was a good thing and that the government debt be retired!
“Well, it all depends, doesn’t it? If the economy is operating below potential because of a demand shock and the government can usefully do something, then it should. But all economies aren’t always in demand-led recessions.”
Maybe/maybe not. But the United States wasn’t obviously running under full potential in 1997. In fact the idea of going into a surplus forced the economy into recession.
The government has the powers to make its expenditure independent of income. In recent times, the US government’s tax revenues fell. Did it automatically induce a reduction in government expenditure?
If government expenditure grows faster than its tax revenue, the process is not sustainable IN THE LONG RUN, i.e., it’s not a stationary equilibrium. That’s what Godley’s model shows. In the short run, different things are possible.
Stable?
Yeah. In fact, government consumption is probably more stable. (I’m going from memory here–I’m sure that it would be easy to find some Burns-Mitchell diagrams online if you’re really interested). But consumption is still pretty stable–more so than output, and much more so than investment or the current account.
are you arguing that income really does not matter?
No, that’s almost the opposite of what I’m arguing. I’m saying that nothing is really exogenous, except in very simple models that shouldn’t be taken too seriously.
Haven’t you studied the multiplier effect?
Of course. But the multiplier is something that commonly features in simple short run models of business cycle fluctuations. It’s not something that is easy to pin down empirically (see the paper that Wh10 posted in the previous thread), and it’s not something that anyone thinks can determine output growth over the long run.
An increase in government expenditure will increase output and due to the multiplier effect, bring in higher tax revenues because of higher economic activity.
Again, this is just an assumption. Why should an increase in government expenditure raise long run economic growth? Perhaps, to paraphrase you slightly, it actually that you think that the government can tell computers how to operate more efficiently? Why does Nigeria have lower per capita GDP than Hong Kong? “Because government expenditure in Nigeria is too low” is not an argument that I would take seriously. Human capital, tangible capital, infrastructure, history, institutions, even genetics are much more plausible places to look.
Also, Tax revenues grow at the same rate as GDP in the long run.
The immediate effect will be to improve the tax inflow at a faster rate than the growth rate.
IF increases in government expenditure cause long run growth, THEN the effect of increasing government expenditure will be to increase income and therefore potential taxes. But this is a pretty weird theory, as far as I can see. Assume that you’re right. Then the government can make the economy arbitrarily large and its inhabitants arbitrarily wealthy just by increasing expenditure. But this is absurd, so…
Not sure! Recently they may have changed. But the atmosphere around the Clinton era was that the surplus was a good thing and that the government debt be retired!
I don’t follow. What does that have to do with it? The reason that I know that neoclassicals don’t always think that the economy is at potential output is because I’m familiar with neoclassical models that try to explain why the economy deviates from potential output.
“If government expenditure grows faster than its tax revenue, the process is not sustainable IN THE LONG RUN, i.e., it’s not a stationary equilibrium. That’s what Godley’s model shows. In the short run, different things are possible.”
General assertions.
Take a closed economy of Godley.
An increase in government expenditure raises output by a multiplier to bring in higher taxes.
On the point about consumption – define stability.
Second there is no natural way in which people automatically start consuming more and this leading to growth.
http://research.stlouisfed.org/fred2/graph/?chart_type=line&s1id=PCE&s1range=10yrs
Stable? Ha!
“No, that’s almost the opposite of what I’m arguing. I’m saying that nothing is really exogenous, except in very simple models that shouldn’t be taken too seriously.”
Well yeah but why did you bring in an autonomous angle in the debate. There is nothing about consumption which makes it automatically grow. Firms need a growth in consumption to expand. Of course an increase in consumer credit can bring this about but we know what happened.
Plus stability is irrelevant to the issue. The problem is that there is no natural way in which consumers consume to produce full employment.
To say that nothing is exogenous is just a simple way of opposing fiscal policy.
“Of course. But the multiplier is something that commonly features in simple short run models of business cycle fluctuations. It’s not something that is easy to pin down empirically (see the paper that Wh10 posted in the previous thread), and it’s not something that anyone thinks can determine output growth over the long run.”
Well because these multipliers are short run multiplier unlike the fiscal stance
“Again, this is just an assumption. Why should an increase in government expenditure raise long run economic growth? Perhaps, to paraphrase you slightly, it actually that you think that the government can tell computers how to operate more efficiently? Why does Nigeria have lower per capita GDP than Hong Kong? “Because government expenditure in Nigeria is too low” is not an argument that I would take seriously. Human capital, tangible capital, infrastructure, history, institutions, even genetics are much more plausible places to look.”
Well, IMF did a mea-culpa recently. It is pure ideology of the economics profession that government expenditures does not raise growth. Paul Krugman writes frequently in this issue. In fact it is his main aim to debunk this hypothesis. Please check all the evidence Krugman provides.
Nations have different fates because ultimately they are balance of payments constrained. You are right – the factors you identify is what determines the income elasticity of imports and exports and ultimately the fate of a nation depends on these. This is the Post Keynesian story of balance of payments constraint growth.
“But this is a pretty weird theory, as far as I can see. Assume that you’re right. Then the government can make the economy arbitrarily large and its inhabitants arbitrarily wealthy just by increasing expenditure. But this is absurd, so…”
Ha!
I am surprised you hold such extreme views.
Plus you are extremising assumptions of the other side. Never did anyone claim that the government can quadruple expenditures overnight. In fact, in Post Keynesian theory, there are explicit statements about constraints – i.e., the capacity to produce, inflationary processes which need political intervention and the balance of payments constraint.
But given the constraints, the Keynesian theory of effective demand still holds. Economies do not have a natural rate of growth. They need active demand management.
I think you may be biased by views of certain sections that fiscal policy can solve all problems. That is an extreme view. But the dogmatic view that fiscal policy is neutral and can only has negative effects is what guides policy all over the world.
An increase in government expenditure raises output by a multiplier to bring in higher taxes.
Again, this is an assumption. You cannot take it as given and use it to prove that government expenditures causes economic growth.
On the point about consumption – define stability.
Same as Godley’s. If the government’s fiscal stance is growing at the same rate as the economy, then the private sector’s fiscal stance must be as well.
Second there is no natural way in which people automatically start consuming more and this leading to growth.
This is no less ridiculous than what you’re claiming here, that if the government automatically starts consuming more, this will lead to growth.
Stable? Ha!
Look at page six here:
http://www2.wiwi.hu-berlin.de/institute/wt2/wintersemester05_06/kb1/KoBeI05_09.pdf
Which shows a Burns-Mitchell diagram for consumption over the cycle for key OECD economies over a 25 year period. Compare it to page four, which shows output.
Well yeah but why did you bring in an autonomous angle in the debate.
You brought it into the debate.
There is nothing about consumption which makes it automatically grow. Firms need a growth in consumption to expand. Of course an increase in consumer credit can bring this about but we know what happened.
Consumption grows because GDP grows. If consumption grows but output does not, then that’s not likely to be a steady state.
Plus stability is irrelevant to the issue.
Stability is central to Godley’s argument.
The problem is that there is no natural way in which consumers consume to produce full employment.
Don’t understand why this should be so, and don’t see what this has to do with what we’re discussing here.
To say that nothing is exogenous is just a simple way of opposing fiscal policy.
Not so.
Well, IMF did a mea-culpa recently. It is pure ideology of the economics profession that government expenditures does not raise growth. Paul Krugman writes frequently in this issue. In fact it is his main aim to debunk this hypothesis. Please check all the evidence Krugman provides.
This is polemical. (It’s not necessary to bring in all these different arguments, BTW. The discussion is complicated enough as it is.) The IMF has been revising its position on the proper response of governments to the financial crisis. It doesn’t think that government expenditure mechanically causes growth, which is what you are arguing. Of course, people want their governments to provide public goods and it is impossible to see how a modern economy could function without an appropriate level of government involvement.
“Again, this is an assumption. You cannot take it as given and use it to prove that government expenditures causes economic growth.”
Ah! Now I get you. You think increased expenditure do not lead to higher growth!
Where did you learn this?
You are a student. Haven’t you learned about Keynesian multipliers?
Plus the graph in the paper actually shows that fiscal policy has had an effect on growth.
Please link me to economics authors who think increased government expenditure do not lead to growth. I suppose Mankiw’s texts also have Keynesian multipliers.
“Same as Godley’s. If the government’s fiscal stance is growing at the same rate as the economy, then the private sector’s fiscal stance must be as well.”
What is a private sector’s fiscal stance????
“This is no less ridiculous than what you’re claiming here, that if the government automatically starts consuming more, this will lead to growth.”
Well, I didn’t say the government “automatically” starts consuming more. It has the power to set its expenditure and a higher expenditure will lead to a rise in output. Now if you do not agree, I see no point in arguing further.
In fact you are being quite unhelpful in the way you are debating. Economists are changing their views and have been saying that fiscal policy does matter.
You on the other hand, think that higher government expenditure does not increase output! Ha ha ha !
“Look at page six here:”
Look at the FRED2 graph!
“Consumption grows because GDP grows. If consumption grows but output does not, then that’s not likely to be a steady state.”
But there is no natural way in which the GDP is growing. It needs demand management.
“his is polemical. (It’s not necessary to bring in all these different arguments, BTW. The discussion is complicated enough as it is.) The IMF has been revising its position on the proper response of governments to the financial crisis. It doesn’t think that government expenditure mechanically causes growth, which is what you are arguing. Of course, people want their governments to provide public goods and it is impossible to see how a modern economy could function without an appropriate level of government involvement.”
I do not know what “mechanically” means. But you want to think that I think the government should spend aimlessly. Seem so from your comment.
Ha!
I am surprised you hold such extreme views.
I don’t see how what I’m saying is extreme. It’s supposed to be a simple logical argument. If the government can set the level of output through a simple multiplier process, then the government can make the economy arbitrarily large. But it’s obvious that the government can’t do this. So it must be the case that the government can’t set the level of output through a simple multiplier process.
Plus you are extremising assumptions of the other side.
Yep. That’s exactly what I’m doing.
Never did anyone claim that the government can quadruple expenditures overnight. In fact, in Post Keynesian theory, there are explicit statements about constraints – i.e., the capacity to produce, inflationary processes which need political intervention and the balance of payments constraint.
So even post Keynesians agree that it’s a lot more complicated than having the government increase expenditure and income increasing as a result.
But given the constraints, the Keynesian theory of effective demand still holds. Economies do not have a natural rate of growth. They need active demand management.
I’m not sure that this makes sense. Let’s say that we get a post Keynesian government and they manage the economy according to whatever principles you like. Then, the economy will grow at a certain rate. Can the government set this rate wherever it likes or is it determined by factors that are beyond its control? I refer you to your answer directly above (starting, “never did anyone…”).
I think you may be biased by views of certain sections that fiscal policy can solve all problems. That is an extreme view. But the dogmatic view that fiscal policy is neutral and can only has negative effects is what guides policy all over the world.
You’ve been arguing that the government can increase growth by increasing its fiscal stance. I don’t think that’s a very plausible explanation of economic growth. I don’t believe that fiscal policy only has negative effects and don’t understand what such things have to do with the discussion here.
Ah! Now I get you. You think increased expenditure do not lead to higher growth!
Where did you learn this?
It’s obvious. Look at the difference in per capita income between any thrd world and first world country. Can you explain this difference with government expenditure? I don’t think so. Take any macro textbook that covers growth models. What are they trying to explain? They’re trying to explain what causes the effectiveness of labour to rise, which is what people think drives the long run growth process.
You are a student. Haven’t you learned about Keynesian multipliers?
Yeah. Like I said, Keynesian multipliers feature in very basic models of the business cycle.
Plus the graph in the paper actually shows that fiscal policy has had an effect on growth.
What the graph shows is that the path of the government’s fiscal stance and GDP are very close.
Please link me to economics authors who think increased government expenditure do not lead to growth. I suppose Mankiw’s texts also have Keynesian multipliers.
Again, the Keynesian multiplier is a feature of simple short run business cycle models. In the long run, growth is meant to be pinned down by supply side factors. This is the standard story in intermediate texts like Mankiw. If you read graduate textbooks, the focus of growth models is on explaining the “accumulation” of the effectiveness of labour, since, empirically, tangible capital accumulation can’t explain cross-sectional or time-series variation in income.
What is a private sector’s fiscal stance????
Well, I put it in scare quotes because you shouldn’t take “fiscal” literally, but it’s easy to create a private sector counterpart to the government’s fiscal stance. Just define private expenditure as P = C + I and the private sector’s share of income as d = 1 – t (so disposable income = dY). Then the private sector counterpart to the fiscal stance is P/d. P/d needs to rise at the same rate as Y if the economy is to be in the steady state. (This will be satisfied if the fiscal stance is growing at the same rate as Y).
It has the power to set its expenditure and a higher expenditure will lead to a rise in output.
I don’t think that it can be the case that raising government expenditure will lead to economic growth in the long run. This is not to say that it might not be able to stabilise output in the short run, or that it doesn’t have desirable properties in any case.
Now if you do not agree, I see no point in arguing further.
No worries!
In fact you are being quite unhelpful in the way you are debating. Economists are changing their views and have been saying that fiscal policy does matter.
Saying that fiscal policy “matters” is very different to saying that government can raise long run growth by raising expenditure. I believe that fiscal policy matters. I do not believe that the government can increase long run economic growth by raising output. I’m more of a short-run person, but I’ve never come across any mainstream theory that argued otherwise.
You on the other hand, think that higher government expenditure does not increase output! Ha ha ha !
That’s correct.
Look at the FRED2 graph!
For some reason, my browser is not happy with FRED at the moment, and I’m afraid I cannot see your graph. But it’s easy to see from the summary Burns-Mitchell diagrams that average the business cycle in major industrial economies over 25 years, that on average consumption is less volatile than output and much less volatile that investment or the current account.
But there is no natural way in which the GDP is growing. It needs demand management.
We know that economies experience fluctuations around long run trends. It seems fair to say that if the government can smooth the cycle, this would be of benefit to society. I don’t see that this means that the government can determine long run growth.
I do not know what “mechanically” means. But you want to think that I think the government should spend aimlessly. Seem so from your comment.
That isn’t what I think. What I think is that long run economic growth is not determined by the level, or the growth in the level, of government expenditures. I personally favour active management of a mixed economy, with high levels of government spending and lots of redistribution–none of which has anything to do with our argument here.
“It’s obvious. Look at the difference in per capita income between any thrd world and first world country. Can you explain this difference with government expenditure? I don’t think so. Take any macro textbook that covers growth models. What are they trying to explain? They’re trying to explain what causes the effectiveness of labour to rise, which is what people think drives the long run growth process.”
I did mention balance-of-payments constraint on growth didn’t I? In fact I more than anyone around in the PKE blogs has raised this point repeatedly.
But you keep claiming that government expenditure is neutral by appealing to the extreme nonsensical case which many people argue that “if that is true, just increase government expenditure. This looks wrong and hence fiscal policy does no good”.
Then you simply reject a basic accounting identity!
I am surprised as to your thinking – given you spend some time in the PKE blogs.
I did mention balance-of-payments constraint on growth didn’t I? In fact I more than anyone around in the PKE blogs has raised this point repeatedly.
Are you saying that differences in the balance of payments can explain cross-sectional and time-series variation in income?
But you keep claiming that government expenditure is neutral by appealing to the extreme nonsensical case which many people argue that “if that is true, just increase government expenditure. This looks wrong and hence fiscal policy does no good”.
I don’t know what you mean by “government expenditure is neutral.” I certainly don’t believe that fiscal policy does no good.
I am arguing that long run economic growth is not a function of government expenditure.
Then you simply reject a basic accounting identity!
No one is rejecting the accounting identity. What I am rejecting is your reasoning, which proceeds from an ex post identity, to conclusions about causality. The government cannot raise private wealth simply by raising its deficit. A real stock does not arise out of a sequence of nominal flows.
Nope.
Haven’t you studied the multiplier effect?
Anyway, this misses the point of my argument, which assumed that government expenditure was growing faster than its tax revenue.
“Anyway, this misses the point of my argument, which assumed that government expenditure was growing faster than its tax revenue.”
Nope doesn’t miss. If you assume that government expenditure is growing faster than the tax revenue, you will obviously get the result that it is growing faster than the tax revenue
If you assume that government expenditure is growing faster than the tax revenue, you will obviously get the result that it is growing faster than the tax revenue
Right, and by Godley’s logic this cannot be a steady state, because eventually tax revenue is going to become arbitrarily small relative to tax revenue.
I agree with vimothy, even if I think Godley may be on to something. Ramanan, you keep making assertions that G/t drives the economy, but the paper does not prove this. The paper provides an expression and interesting charts, which I think vimothy has shown have much to answer for. It does not provide a rigorous proof for the causal relationships you are asserting.
Ramanan, I am not saying your intuition is wrong, but you are avoiding vimothy’s argument. His point is that G/t, while not equal to GDP, will track it. So your response that “Look (G + X)/(t+mu) is not equal to GDP. So your analogy is not right” misses the point, I think.
wh10,
Vimothy is trivializing the issue. He is claiming the graph just proves Y=Y
I don’t think that’s trivializing. He is critiquing the methodology and pointing out that the expression runs the risk of naturally tracking Y. That is potentially problematic.
“I don’t think that’s trivializing. He is critiquing the methodology and pointing out that the expression runs the risk of naturally tracking Y. That is potentially problematic.”
It is because it output naturally tracks the augmented fiscal stance
Of course, that is your conviction, but it doesn’t address his critique.
Vimothy is confusing things.
In the medium run – *according to the model* – G/t will equal Y (if the government expenditure for example is increased from G_0 to G).
Hence he thinks that in the paper Y is set to Y.
But is (G+X)/(t+mu) equal to Y as an identity?
No.
(G+X)/(t+mu) can be anything. It can be 1/2 output. It can be 2 times output. It can be 0.75 times output. There is nothing in that expression which naturally makes it equal to Y.
Let me give an example and this is not an exaggeration.
If I have a behavioural model which assumes Ricardian equivalence, then there is no effect of an increase in G.
If I were to write a model with this assumption,
(G+X)/(t + mu) will not be equal to Y in the model.
“My memory of Godley’s paper (which I scanned) is that he not actually claiming that deficits are essential for growth, but that for the economy to grow, the government’s fiscal stance has to rise.”
I think this is actually about medium of exchange. The amount of medium of exchange in circulation is a commodity that trades relative to the amount of goods/services (one thing).
“That also doesn’t make sense to me. When I think about what determines growth, I think of things like the accumulation of human capital, …”
That sounds like real AS? What about real AD?
Okay, so, vimothy may appreciate this from me.
I actually did not find the Godley paper to be that enlightening (yet). He seems to pull an equation out of thin air and provide minimal theoretical support for it. The regressions look compelling on the surface, but I am still not sure what I am looking at.
As for the equation, GDP = G/t only holds if G=T. What does “G/t” even mean when G does not equal T? And how does it therefore follow that for the economy to expand, “G/t” should rise? Vimothy’s “multicolinearity” point seems potentially relevant.
Additionally, as a theoretical matter, why can’t inside money creation drive growth in this model? Godley didn’t build Minsky into it, so as to make such a path too unstable.
For multicollinearity, see my comment above.
G and t are exogenous – under the control of the government. X and mu are trade performance parameters and again no collinearity problem.
G/t appears in all theory papers by Godley where he actually derives it rather than pulling it out of thin air. (see ref to his own book Macroeconomics). Also other Keynesian authors also have this expression in their work.
The main point is that economies have been driven by fiscal policy and trade performance with private expenditure just responding to it. See the reference to differences in the graphs explainable by private sector borrowing.
Of course if private expenditure increases relative to income, this can also drive growth but it cannot continue forever because the private sector’s indebtedness keeps rising and this has to give in. Given this, if the fiscal stance is not adjusted, the economy will run into a recession.
So the US economy did head into recession and was again reflated by the strong fiscal expansion of the Bush administration!
“What does “G/t” even mean when G does not equal T?”
If G is $3T and t is 20%, G/t is $15tn.
“And how does it therefore follow that for the economy to expand, “G/t” should rise?”
Because in Keynesian models, the economy responds to the fiscal stance of the government. output is dependent on fiscal policy.
““What does “G/t” even mean when G does not equal T?”
If G is $3T and t is 20%, G/t is $15tn.”
I realize that, but I don’t understand what G/t is supposed to mean, other than something Godley calls the “fiscal stance.” It means GDP when G=T, but otherwise it just seems to be one random variable divided by another. I understand the argument that G and T are important for growth, but I don’t understand how this expression elucidates that. Perhaps I need to see the derivations in his other books.
““And how does it therefore follow that for the economy to expand, “G/t” should rise?”
Because in Keynesian models, the economy responds to the fiscal stance of the government. output is dependent on fiscal policy.”
He doesn’t prove that though! Michael set my expectations up such that I thought he would prove this with math and theory. Instead, he asserts it, creates some random expression, and then makes some compelling plots.
“He doesn’t prove that though! ”
Wh10,
The actual proof is not straightforward. So left to references. It is not only him who has this expression but other Keynesian authors too!
However there is sufficient intuition provided in the paper which is Carl Christ’s back of the envelop calculation. If an economy is in a state and the government expenditure is suddenly increased, we need to find what the new level of output will be. In a steady state (non-growing), inflows will be equal to outflows and it is trivial to get the output since G will be equal to T and since T = tY, Y will be equal to G/t in the future (not now). Now G is not equal to T and G/t is not equal to Y.
(Not the best analogy: Does NASA always repeat the proof of Kepler’s laws?)
This is theory. In real life government expenditure is increasing and there is no steady state with a balanced budget. But it does give an intuition of the kind If government expenditure is increasing at a growth rate g, G/t will track Y and both will grow at the same rate.
“The steady state case is just a quick way of getting what a good definition of fiscal stance is.”
This is my point. The “fiscal stance” is only anything when at steady state; at steady state, it is equal to GDP. When not at steady state, I have no idea what it means – it’s just two macro accounting variables divided together. Furthermore, I don’t see how the knowledge that “if ALL flows go to zero, the stocks can’t grow,” means that G and t need to increase/decrease for GDP to grow in the future. The former does not establish any specific causal relationship on G/t for GDP outside of G=T. But Godley makes this leap in the paper, without explanation– “It would then follow that if the economy were moving toward stock-flow equilibrium and if taxes were levied as a proportion of income, the GDP of a (closed) economy would always be tracking, perhaps with a long lag, government outlays divided by the average tax rate – the very same concept that we call fiscal stance. Therefore, a necessary condition for the expansion of the economy, at least in the long term, is that the fiscal stance should rise: Government expenditure must rise relative to the average tax rate.”
As for the rest of your comment, this seems to be the arguing and modeling Godley should have been doing the paper to support his proposition, but it is all absent. As I said below, I expected rigorous proof.
Sorry, this was in reply to Ramanan at 8:34AM.
A complete proof cannot be provided in an article such as this. One cannot keep writing complete proofs everytime – especially given that it takes a lot of space equivalent of a book.
But references are provided. Can you please check his book Macroeconomics?
On the other hand, the proof is the graph. The fiscal stance or more correctly the augmented fiscal stance actually comes close to it.
“I have no idea what it means – it’s just two macro accounting variables divided together. Furthermore, I don’t see how the knowledge that “if ALL flows go to zero, the stocks can’t grow,” means that G and t need to increase/decrease for GDP to grow in the future. ”
If the government sets its expenditure at G and average tax rates at t, then the economy will move toward a state in which the GDP will be G/t (simple closed economy model). It won’t move immediately because it takes time for the policy to have an effect.
The underlying idea is that the budget deficit is a poor measure of a policy stance. Take the case in the United States. The budget deficit is high – does that mean that fiscal policy is highly expansionary?
No.
So one needs another measure to see if fiscal policy is expansionary or contractionary. The fiscal stance measures that. It has the right look. Higher expenditures increase demand and lower taxes also.
Now one can construct zillions of things which have this property so we need some kind of argument to judge which one is more appropriate.
So G/t is chosen as the closest proxy because if G is increased, output will likely rise and if it is set permanently at this new level, output will rise to G/t.
(That is in the closed economy case)
Of course there are many assumptions – other things unchanged etc – economics is not an exact science but that shouldn’t prevent one from arguing the importance of something.
Think about t = T/Y as being the government’s share of output. Define another variable called the private sector’s share of output (= 1 – t). Imagine that the growth of Y is constant at whatever rate. A stable equilibrium has the government’s expenditure and the private sector’s expenditure growing at the same rate as their respective incomes. Right? Think about the implications of that not being the case. If your expenditure is growing faster than your income, that’s ultimately unsustainable and in the long run something has to give. That’s what the steady-state equilibrium tells you: it’s not sustainable for expenditure and income to grow at different rates, because as time becomes very large, this process will explode.
In Godley’s models, there are sustainable processes and unsustainable processes.
If the government increases its expenditure, it will temporarily raise the budget deficit but a higher output and the growth in the output will bring in sufficient taxes to prevent the deficit from exploding.
In a closed economy growth model, the government will hit primary surpluses without actually attempting to go into that surplus.
But that’s simply an assumption of the model. Here we are interested in seeing to what extent it’s actually true by examining some evidence.
Evidence for?
Not necessary to actually think in terms of models.
There are unsustainable processes now (but that is due to global imbalances and let me not get into it now and just take the case of a closed economy).
Let us suppose that the government’s budget balance keeps increasing. As a mirror, the private sector’s surplus and as a result wealth is also increasing.
Unless the private sector is adamant with a “Not consume till I die attitude”, why will they not spend?
Let us suppose that the government’s budget balance keeps increasing. As a mirror, the private sector’s surplus and as a result wealth is also increasing.
[Alarm bells start ringing!]
You CANNOT reason from ex post identities in that fashion. Bad, bad Ramanan. The private sector’s wealth is NOT increasing as a result of the government’s budget balance increasing. That is a terrible argument. As penance, you are required to read Paul Krugman’s blog every day until you’re sorry and/or understand why it’s a terrible argument.
Okay I meant the deficit not the surplus, but that was not your point anyway right?
“The private sector’s wealth is NOT increasing as a result of the government’s budget balance increasing. That is a terrible argument. ”
Yes, it is increasing if the government’s deficit is growing.
Not sure why you are ringing alarm bells.
[Alarm bells still ringing! Lights start flashing!]
The government cannot raise private sector wealth simply by increasing its deficit. Nuh-uh. That’s reasoning by identity and it’s very bad–the sort of stuff that Krugman hammers Chicago types for with reasonable frequency.
Private sector Net Accumulation of Financial Assets = Govt deficit.
Haven’t you heard this identity
I thought you hang around PKE blogs a lot and must be aware of this.
Plus secondly, the government sets its policy i.e., tax rate and expenditures. The deficit depends on the activity of the private sector and how it responds to stimulus etc.
Nontheless
NAFA = Govt Deficit.
Yes, I hang around PKE blogs and I’m very familiar with this identity, which is an identity, not a causal relationship. It is a category error to confuse the two. This confirms the diagnosis and you are required by the terms of your licence to read Krugman’s blog until you understand this.
But you directly claim that if the budget is in deficit, it doesn’t increase the private sector deficit.
You claimed.
Let me quote you
“The private sector’s wealth is NOT increasing as a result of the government’s budget balance increasing. That is a terrible argument.”
Which means you are claiming that if the government’s budget deficit is increasing, the private sector’s wealth is not increasing.
Look who is confused?
Krugman struggles to understand simple monetary theory matters.
It is you who is committing the error.
Plus it is the easiest thing to verify this.
During the financial crisis, the United States’ government has run huge deficits. At the same time, the private sector’s wealth (both financial and nonfinancial) has been increasing.
But – you reject this identity.
It has also had the beneficial effect of helping the private sector deleverage.
But you directly claim that if the budget is in deficit, it doesn’t increase the private sector deficit.
I am not saying that NAFA != Govt deficit, but that the government cannot increase PRIVATE WEALTH by increasing its deficit.
Take the GDP identity. GDP = C + I + G. Then, to increase GDP, simply increase G. But this is reasoning from an ex post identity that is necessarily satisfied in all states of the world to a causal relationship between nominal government expenditure and real income. It’s faulty logic.
Which means you are claiming that if the government’s budget deficit is increasing, the private sector’s wealth is not increasing.
Private wealth might be increasing or it might not, but the government cannot simply raise private wealth just because you have an ex post relationship that says NAFA = the deficit.
Krugman struggles to understand simple monetary theory matters.
Krugman is not a monetary economist, but he generally understands the economic significance of accounting identities quite well.
“I don’t see how what I’m saying is extreme. It’s supposed to be a simple logical argument. If the government can set the level of output through a simple multiplier process, then the government can make the economy arbitrarily large. But it’s obvious that the government can’t do this. So it must be the case that the government can’t set the level of output through a simple multiplier process.”
You keep missing the point. I have repeated four-five times that it is not possible for the government to quadruple its expenditure overnight.
Then you raise alarm bells and claim that deficit doesn’t increase private sector wealth!
The point is that there is simply enough room for governments to increase expenditures and raise output before supply constraints are hit. There are of course exceptions that you may find in some places. But in general these are always exceptions.
Human capital etc are important for economic growth but economies by themselves are not guided by an invisible hand to naturally exploit resources and left to themselves stagnate and/or end in financial crises.
Second supply constraints are not exogenous as neoclassicals tend to think with their production functions. A higher growth of output also leads to firms increasing their capacity in advance of sales etc but let me not get into that.
But, you don’t seem to follow this point at all
It is a bit like a car. Every car has an upper limit on speed but it needs a driver to move it and steer it. Most economists tend to view the economy as a car with no driver.
Vim,
Sorry starting a separate thread. Difficult to follow otherwise:
“I am not saying that NAFA != Govt deficit, but that the government cannot increase PRIVATE WEALTH by increasing its deficit.”
That is not right. A good part of wealth of private sector of nations is due to the public debt.
Change in Private Sector Wealth = Govt Deficit + Increase in Nonfinancial assets + Revaluations.
Now, there is some sort of argument in your comments and you came back later claiming that:
“No one is rejecting the accounting identity. What I am rejecting is your reasoning, which proceeds from an ex post identity, to conclusions about causality. The government cannot raise private wealth simply by raising its deficit. A real stock does not arise out of a sequence of nominal flows.”
First, you claim total impossibility. There is nothing preventing a government from inflating away and thereby reducing the real wealth. But this is not always the case. Again you talk extremes – a bit like an Austrian would argue.
If you do inflation accounting, you will find that not only are things deflated by the price but there is another term for real wealth which is the erosion due to price rise (with a negative sign).
But it certainly does not follow that any deficit will lead to the erosion term being greater than the other terms in the accounting equation.
In recent times, in the United States, both real and nominal wealth have risen because of the deficit.
“Are you saying that differences in the balance of payments can explain cross-sectional and time-series variation in income?”
Yes, in the world of free trade – the one in which we live – the growth of nations finally depends on how successful their producers are in international markets. So what is say is of course true – investment in innovation, research, education are all important. The growth is dependent on the growth of exports and inversely to the income elasticity of imports. The elasticity captures all the supply side things.
So a poor nation won’t become Hong Kong by fiscal policy alone.
If the government relaxes fiscal policy, not only will it meet with capacity constraints (and pressures due to inflation) but also a balance of payments constraint – meaning fiscal policy will have to give in. (If not – the way you suggest – public debt and external debt will keep rising relative to gdp, as you suggest. The reason I object to your arguments is that you suggest that in all cases). But that again doesn’t put an explicit constraint because foreigners can be attracted to invest – such as via FDI but ultimately fiscal policy gives in and a lot of nations suffer the fate.
But none of this implies that fiscal policy has the kind of limited role you are talking. Economies still are unable to exploit the full usage of improvement of the supply side and fiscal policy has a place of its own. Nations do not make use of the space available to them.
My issues with you is that you are extremising the assumptions.
Right now, very few nations can actually do a unilateral fiscal expansion. This is the reason one hears of international coordination of policy but the bias against fiscal policy is so much that it seems a difficult task politically.
That is not right. A good part of wealth of private sector of nations is due to the public debt.
Yes, but this does not mean that the government can simply increase private wealth by increasing its deficit any more than the equality of national income and aggregate expenditure means that you can simply increase national income by increasing aggregate expenditure. This is very flawed reasoning that relies on the fact that identities are always satisfied after the fact. What identities don’t tell you is what is going to have to happen in order to reconcile all of the ex ante “plans” of agents in the economy. All you have is an identity that says NAFA = deficit. Given that NAFA = deficit, if the deficit increases, then so too does NAFA, but this does not automatically mean that private wealth increases. It might increase, or it might decrease, or it might stay the same. It’s not possible to say simply from your ex post identity.
First, you claim total impossibility. There is nothing preventing a government from inflating away and thereby reducing the real wealth. But this is not always the case. Again you talk extremes – a bit like an Austrian would argue.
You have the extreme arguments here. You have claimed that fiscal policy can determine long run economic growth, and you have claimed that the equality of NAFA and the government’s deficit means that the government can raise private wealth by increasing the deficit. (It’s easy to conceive of ways in which the government can affect private wealth, but these rely on a lot more than a simple accounting tautology).
But it certainly does not follow that any deficit will lead to the erosion term being greater than the other terms in the accounting equation.
Nothing follows from the identity other than whatever happens, the identity will be satisfied. That’s all she wrote. Other than that–nada, zip, zilch, nothing.
Private wealth measures the future real income of the private sector. If the government increases the level of its nominal expenditures, will this measure go up or down? Obviously, it depends. But you cannot say that it will rise simply because NAFA = the govt deficit–that is a fallacious argument.
In recent times, in the United States, both real and nominal wealth have risen because of the deficit.
What? Because of the deficit? Because–how?
Yes, in the world of free trade – the one in which we live – the growth of nations finally depends on how successful their producers are in international markets.
And what determines how successful their producers are in international markets?
So what is say is of course true – investment in innovation, research, education are all important.
Right!
The growth is dependent on the growth of exports and inversely to the income elasticity of imports. The elasticity captures all the supply side things.
This is a proximate argument. Something determines elasticities, which are just summary statistics.
So a poor nation won’t become Hong Kong by fiscal policy alone.
That’s a bit of an understatement!
But none of this implies that fiscal policy has the kind of limited role you are talking.
I am not advocating for a limited role for fiscal policy, except to say that fiscal policy cannot determine long run growth.
Economies still are unable to exploit the full usage of improvement of the supply side and fiscal policy has a place of its own. Nations do not make use of the space available to them.
I certainly think it would be better if there were more redistributive policies and progressive tax systems. I also think that fiscal policy has a role to play in stabilising short run output fluctuations. But we’re not talking about every possible facet of fiscal policy. We’re talking about the effect of fiscal policy on long run growth.
My issues with you is that you are extremising the assumptions.
If you follow your assumptions to their logical conclusions and you find yourself somewhere ridiculous, then what do you do? You question your assumptions.
“Given that NAFA = deficit, if the deficit increases, then so too does NAFA, but this does not automatically mean that private wealth increases.”
Is the issue here that we need to distinguish between nominal and real? An increase in nongovt net financial assets (which what a deficit does) IS by definition an increase in NOMINAL private wealth. I don’t see how this is debatable.
Yes, this seems to be a definitional issue. I’d be curious in JKH’s thoughts here. Vimothy, you write:
“Private wealth measures the future real income of the private sector”
So embedded in your definition is that we’re taking the present value of *future* *real* income. As an analogy, this seems similar to how an investor would value shareholder equity in a company, taking the present value of future cash flow to equity.
This is not the definition commonly used in the PKE blogosphere, which seems to be more along the lines of: private sector *nominal* wealth = nominal assets – nominal liabilities. Additionally, this definition seems to use something like a ‘book value’ method of valuation. There is no mystery to these numbers, no need to predict project future cash flow or anything like that. You can easily find them using the national accounts. They’re the sectoral balances (http://pragcap.com/updated-sectoral-balances)
. Using that definition, yes, a higher deficits does indeed mean increased private sector nominal wealth.
Vimothy’s definition complicates things, and I think he is right, generally, if we’re using his definition. But both definitions seem to be useful lenses to look through, depending on the situation, IMO.
Let me take this a step further, and again, I would love JKH’s thoughts. This is right up his alley, I think.
To clarify, the reason it complicates things is that it requires us to predict the future, since we’re taking a present value of future income. In that case, of course we can’t reason from an accounting identity that only defines the here and now in terms of book value.
But I want to make a careful distinction here. It’s not the methodology of accounting, in general, that is limiting here, nor is it that we’re looking ‘ex post’ (after the fact of an event). It is that, I think, we’re using a book value method of accounting – the kind that gives us the sectoral balances that we’re familiar with. If we use ‘mark to market’ accounting, then our accounting identity requires us to predict the future and discount the relevant data to present value. Then our accounting identity does what vimothy wants it to do, and it is still ex post, in the sense that ‘ex post’ means ‘after the fact.’ But this is very different than the ‘book value’ sectoral balances we’re used to talking about.
(Crap, should have said at the outset that I want to put the concept of real vs nominal aside. That also complicates things, of course. Again, our accounting identities can deal with that as long we measure things in terms of real values, but it requires us to model and predict how inflation will change etc.)
Vimothy, this is all agreed upon. You really should modify your use of the word “wealth” to be more precise, though. You should say “real wealth,” because “nominal wealth” exists, too, even if you think it’s not as useful a concept, in general.
Sorry^, that was in reply to Vimothy’s post at 8:21 pm. Also, vimothy, have you taken any finance or accounting courses? Your assumption that “wealth” means not only “real wealth” but is also a measure of the present value of future real wealth is not at all a universal default definition. I don’t care how much more informative that definition is, it’s just not the universal default definition.
Nope, never taken any accounting courses.
I suppose I’m taking it as given that since we’re discussing macroeconomics, wealth is a real variable. When you are discussing your tax returns with your accountant, I don’t doubt that this is academic, but the idea that society as a whole can be made to become more wealthy *in purely nominal terms* seems like an unimpressive feat.
But I will try to be more careful about making this distinction in the future.
I agree, just saying
Okay, maybe I could try to separate two things that have become entangled here: why it’s bad to draw conclusions about causality from accounting identities and why, if the government increases its deficit, even though the private sector necessarily ends up holding more net assets, cet par, it does not necessarily have higher wealth.
An identity is a tautology, It cannot be wrong. It is satisfied in all states of the world. Whether Keynes or Mises was right about everything, NAFA will always equal FIFL, Y will always equal AE and so on. The meaning of “ex post” is precisely that whatever happens, GDP = C + I + G + NX. If the government tries to raise GDP by raising G, it might cause GDP to rise, or fall, or stay the same. Whatever actually happens, we will end up with GDP = C + I + G + NX. In order to get from GDP = C + I + G + NX and increase in G to increase in GDP you have to assume that if you raise G, GDP will also rise. This might be a good assumption, or it might be bad, depending on the context, but it certainly doesn’t follow from the existence of the identity itself.
Now, the private sector’s acquisition of net assets is just the flip side of the government’s issuance of net liabilities. Given that the government has issued assets, it must be the case that the non-government sector has acquired assets. They’re one and the same thing (that is, they reference the same thing in two different senses). If the government issues a bunch of bonds, the private sector doesn’t become more wealthy in toto unless its income in the future is higher as a result of the bonds being issued. But its income in the future will not be higher unless issuing the bonds somehow makes it more productive. And how do bonds do that? But if we have to ask the question then we’ve already arrived at the point at which we can see that we cannot raise private wealth simply by issuing government liabilities. If only!
Wh10, No one is debating it, as far as I can tell. “Nominal private wealth” is just a number. It doesn’t have any particular significance if and of itself. Jf the nominal value of wealth doubles, but the real value of wealth does not change, what does it matter? Is it better for the nominal wealth of the private sector to be a hundred units, two hundred units, or a thousand units?
If we call the government deficit the “net issuance of financial liabilities,” then by identity we have NAFA = NIFL. If the government issues more net financial liabilities, then for sure the non-government must acquire more net financial assets. But this does not mean private wealth has increased. It just means that the government has issued some liabilities, and therefore that someone must be holding them.
Agree. See my above posts. And it’s not just the distinction between nominal and real. It’s also whether or not the future is “in scope,” and we’re dealing with discounting to net present value. Your definition invokes both issues according to what I quote.
That’s right. But it’s not necessary to be able to comprehensively account for wealth or the future path of income to be able to see that, in and of itself, issuing some extra bonds will do very little change the private sectors ability to produce more output in the future.
The government might be able to change the private sector’s ability to produce income in the future, of course; for example, by building more primary schools, or declaring war on its neighbours–but these do not follow from the fact that the government has the power to issue securities.
Incidentally, you might enjoy the following paper by Willem Buiter: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=304754
Oh cool – thanks. This seems to get at what I was trying to explore above. I did a bad job of it though. Deficit = nafa from a “book value” perspective, but if we change to “mark to market” and discount wealth to the present, then we don’t have that identity anymore. (just ignore me if I am not making sense.)
Advantage Ramanan!
http://www.dailymotion.com/video/xapin_advantage-agassi_fun
What data do we need to pull from FRED the last few quarters to track fiscal stance for the last few quarters?
GDP, G and the government’s share of income.
I am headed out so a limited reply.
Vimothy is quite wrong. The Buiter paper he quotes is not actually implemented by national accountants. It is based on the crude and dangerous idea that deficits do not increase wealth.
Please check Flow of Funds.
It is the general denial by the profession that government debt is not wealth for the private sector. Many neoclassical economists have written on this.
Okay, haven’t looked at Buiter yet.
But whatever Buiter says, Vimothy’s accounting is very wrong. He says deficit does not increase the wealth of the private sector.
Then he raised alarm bells twice! First he gets the nominal accounting wrong and then tries to present a Monetarist story.
You have browsed MMT sites and that should come out as wrong to you. But you can confirm with them.
There was a blog by Bill Mitchell that even though some neoclassical economists were aware of this from the 70s, they tried to argue that it is really not. Search Barro in his blog (dont have links).
There main mechanism suggested in such works is that the government supposedly monetizes its deficits and this wipes out nominal wealth increase by eroding the real value of the wealth – a totally Monetarist story.
Then he raised alarm bells twice! First he gets the nominal accounting wrong and then tries to present a Monetarist story.
1, I am not disputing the “nominal accounting.” I am disputing your claim that the government can raise private wealth simply by issuing liabilities. Of course, I agree that the government can actually issue liabilities and that what it issues equals what people acquire. That’s fine, but so what? The argument is over whether, merely by virtue of the fact that the government issues liabilities, there is some tangible effect on the future productive capacity of the economy.
Why don’t you explain to me how it works. The economy is in whatever state. Then, the government increases the supply of bonds at the margin. Then, there is [some kind of effect]. Then, the economy’s ability to produce is higher in the future. What is [some kind of effect]? Take me through it. Is the relationship between deficit and wealth one-for-one, sloped, curved? Does it only work for the government, or does it work for any other sectors or agents?
2, The “monetarist story” is another polemical distraction. I am not a monetarist, and I’m not sure how it is relevant because we haven’t discussed monetarism in any way shape or form. What we’ve discussed is two ideas: whether the government can increase economic growth by increasing expenditure; and whether the government can increase private wealth by increasing its deficit.
There main mechanism suggested in such works is that the government supposedly monetizes its deficits and this wipes out nominal wealth increase by eroding the real value of the wealth – a totally Monetarist story.
Ricardian equivalence isn’t a totally Monetarist story. It doesn’t really have anything to do with monetarism. Ricardian equivalence says that the private sector does not *treat* govt bonds as net wealth. It doesn’t argue about whether they *are* net wealth. If the private sector doesn’t treat govt bonds as net wealth, then the time path of taxes doesn’t matter. But this has nothing to do with monetarism and nothing to do with the discussion here.
Throughout this discussion, Ramanan has tried to tie my arguments to some wider polemical point he’s trying to make. Neoclassical economists don’t think that fiscal policy can do any good, the IMF did this when it should have done that, and now–all economists believe in Ricardian equivalence. This doesn’t have anything to do with our argument.
Let’s not get side tracked by Buiter either. That is a tangent. If you’re interested in thinking about public and private balance sheets in a comprehensive sense that tries to take into account implicit and nonmarketable assets, Buiter is one way to go. Combined with the contingent claims lit from Gray and co, I have found it to be quite a helpful perspective on the interplay between sector balance sheets and financial crises. Anyway.
The idea that deficits do not increase wealth is not crude. What’s crude is the argument from accounting identity that goes: since net acquisition of financial assets for the private sector equals net issuance of financial liabilities for the public sector, to increase wealth, increase net issuance of private liabilities.
This is, I’m afraid to say, a very bad argument that exploits the fact that accounting identities are always satisfied after the fact to argue it is possible for the government to cause wealth to be higher, merely by issuing liabilities! If this is so, given that the government can supply infinite liabilities, it can also supply infinite wealth. So why shouldn’t it? If this is so, the government can supply infinite GDP, simply by increasing G. So why doesn’t it?
Because it’s impossible. For the government to raise the wealth of the private sector, it needs to raise its income in the future. So how does issuing a few bonds at the margin achieve that? Obviously, it does not. If increasing the deficit were to increase private wealth, then it must be the case that the deficit was tied to some tangible effect on the economic resources that the economy can command. It cannot fall simply out of the fact that deficit = net asset acquisition.
Divide the economy into me + everyone else. Then I raise the net wealth of everyone else by borrowing a million. Brilliant! But why stop there? I could borrow more: two million, say or current GDP. But this is silly.
You err again.
You still claim that deficits do not add to private wealth.
You try to prove this by appealing to extreme situations and raise alarm bells.
You even try to prove this by using net incurrence of liabilities and all that. But that is a shoddy piece of accounting.
Please see numerous blogs where both MR and MMTers show the accounting.
relaxing fiscal policy slightly before inflationary pressures from demand starts hitting is quite different from the Austrian scenario you present.
Your argument is a bit like saying what is the use of medicine? If taking medicines can be good why not popup zillions of medicines and we can live forever.
Its is you who is being crude here.
Ramanan,
Are you following my argument?
I am not disputing the identity. As I’ve said (repeatedly), the identities must always be satisfied. They are logical tautologies. In every state of the world, non-govt balance = govt balance.
You cannot understand the interplay of government borrowing, expenditure and output using simple aggregate accounting identity like the one above. You need *economic theory.*
You say that the government can increase private wealth simply by increasing the deficit, but the only argument that you’ve made so far has been purely semantic: if we define private wealth as govt liabilities plus some other stuff, and if the govt increases the supply of its liabilities, the private sector must have more wealth. Equally, if we were to somehow double every price, including all debts and contracts, in the world overnight, everyone would be twice as wealthy.
But more marginal bonds for the private sector doesn’t actually increase its wealth as a whole unless it increases its future income as a whole. And how do marginal bonds do this? If you have to ask, it must be the case that you need something else other than the identity to explain it.
You say that I am going to extremes and that you don’t actually mean anything like this, but what do you mean? Why don’t you qualify your incredibly broad statements? If the government can only increase private wealth *up to a point* by increasing its deficit, then what is that point? If it tries to double private wealth, will it only increase by 80 percent, or 75 percent, or perhaps only 50? Does it depend on the level of prior wealth? Is it contingent on particular states of the world? If so, what are the states? What is the actual mechanism that goes from the government making a promise to provide some future economic resource to someone to there being more economic resources in general from which to obtain it?
vimothy,
I highly recommend you read the S = I + (S-I) posts and comments.
I am missing the point you are trying to make because at you recognize the government can raise real world wealth in some (or even many) situations.
But you’re denying exactly what? Nobody here is claiming the government can raise output to any level just by spending more.
What I am saying is there are real world constraints on growth, and the only reasonable way to put these into model would be to start from the Godley matrix. If you start anywhere else, you end up with papers like Cebula’s paper, where he’s not measuring the right thing, mixing up real and nominal, and then make sweeping conclusions which are dangerously wrong.
The identities are a feature not a bug. You’re thinking about a category error between causation and identity which you should be exploiting, not dismissing.
We don’t have good ways to measure real wealth. Our methods are terribly crude at best, despite the enormous implications of increasing our real wealth.
So while I understand your concerns – again please look at the S = I + (S-I) debates for more – what I don’t understand is why you’d want to try to raise the valid question “Can the government increase wealth?” in this manner. I don’t understand why you’d want to portray Ram – or us here at MR – as people who believe we can become gods overnight by simply increasing government spending.
The end result of increasing real wealth is to become something like gods, where dreams become reality at a thought, because our control over matter and energy become that refined.
What we can do is decompose GDP (or GDP relationships) into accounting compliant identities which make our determination of the components of real growth more accurate.
Because as far as I know, nominal GDP is one of the most accurate methods of determining real GDP we have available to humans.
Right now, we’re having a terrible, terrible time simply convincing the wider population we can create real growth by having the government spend more money right now, today, as I sit and write this.
My personal take is the government is possibly the single largest driver of human wealth. Not that it creates wealth, but rather it provides the framework in which creating real wealth either makes sense or not. It doesn’t make sense to try to make your life better with new goods if you’re in danger of getting killed or cannot profit from your talents.
Then, private investment makes the biggest input. But jeez, having some sort of reasonable discussion about how best to create real wealth cannot start with claiming the MR crowd thinks the world can become gods by increasing government spending 100-fold in the next year.
Mike,
I read many of the S = I + (S – I) posts at the time, and commented on some of them. I don’t see what they have to do with the discussion here. Perhaps you could elaborate.
Ramanan has argued that, BECAUSE net financial assets equals net financial liabilities, the government can raise private wealth by increasing NFA. This is a semantic argument, not an economic argument. The identity of NFA and NFL is not sufficient. You need the tangible effects on economic resources that raise future potential output, which I have been discussing. Just because NFA = NFL doesn’t mean that the government can raise private wealth by just issuing some more bonds. It can try, and it might succeed, or it might not. It is hard to say without saying more. But if it succeeds it will not be because NFA = NFL.
Ramanan has also argued that, BECAUSE government revenue and expenditure must grow at the same rate to ensure a steady state equilibrium, the government can increase its expenditure and its revenue will increase, which means that if it holds its share of income constant, the level of income must rise to ensure long run equilibrium. This is also mistaken. Again, the causality is backwards. Since any outcome other than the fiscal stance and GDP growing at the same rate is unsustainable, you will not see it happen, as rule.
But this means that it’s impossible to use this fact to select between alternative explanations. Here’s an example hypothesis: If the fiscal stance ever grows faster than the economy, government debt will eventually become very large relative to the economy. Therefore policy makers will not, and in fact cannot, let such a situation persist for any length of time; they will either lower their expenditure, or raise their share of income. In the long run, we will see that the fiscal stance grows at the same rate as the economy.
There is nothing in the series in Godley’s paper that will help you select between the two hypotheses, since they both predict the same outcome, which we observe. Moreover, we all know that government’s really do try to raise taxes or lower their spending when they feel that their debt is becoming too large. The government here (the UK) has spent the last few years doing little else.
Cullen has a timely post – http://pragcap.com/japan-corporate-profits-kalecki-and-living-standards
I really do think we’re all on the same page here. vimothy, I don’t think Ramanan is committing the mistakes you believe he is. There is a some word-twisting / taking statements too literally going on here. His thinking that G can play a role in long-term growth doesn’t mean that he thinks increased nominal wealth ALWAYS equals increased real wealth, but he does see ways that it can, even in the long-term.
I distresses me that I’m such a bad writer that I can’t even convey my argument to the other commenters here, let alone convince them of it.
Let me try to say this more plainly.
Ramanan has made two arguments:
A1, BECAUSE, in the long run, the growth in the government’s fiscal stance must equal the growth of output to be an equilibrium, THEREFORE the government can raise output by raising expenditure relative to its share of aggregate income.
A2, BECAUSE private wealth is private assets less liabilities, and BECAUSE net financial assets equals net financial liabilities, THEREFORE the government can raise private wealth by issuing more liabilities.
These are both BAD ARGUMENTS.
I’m not saying is that the government can’t meaningfully contribute to the economy, or affect social welfare. I am not saying that the government cannot raise or lower wealth or raise or lower output. (In some circumstances clearly it can, and in others, it cannot). I’m not saying that fiscal policy does no good and I’m not saying that fiscal policy is neutral.
I’m saying that the two arguments above are invalid.
A2 is nonsensical.
I claim it is the task of the government to raise demand and this works via the multiplier process and I really think supply side contraints are overblown and that demand creates its own supply. Of course if demand is increased fast, supply contraints will be hit and this puts a constraint on policy.
A2 was you thinking that I think because the government deficit increases wealth it should increase spending. No!
I gave the example in a situation of a closed economy (hypothetical) where the deficit continues to rise. But this is not possible. Increased injection will lead to higher and higher wealth but before such situation is reached, there will be higher private expenditure and more tax inflows – preventing an explosion of debt/gdp.
So it is highly unlikely unless one is talking of extreme cases.
Of course with the qualification that there is also a balance of payments constraint which is highly important in my view.
Ramanan,
Here’s the original exchange, edited for brevity:
RAMANAN: “Let us suppose that the government’s budget balance keeps increasing. As a mirror, the private sector’s surplus and as a result wealth is also increasing.”
VIMOTHY: “The private sector’s wealth is NOT increasing as a result of the government’s budget balance increasing. That is a terrible argument.”
RAMANAN: “Yes, it is increasing if the government’s deficit is growing.”
VIMOTHY: “The government cannot raise private sector wealth simply by increasing its deficit. Nuh-uh. ”
RAMANAN: “Private sector Net Accumulation of Financial Assets = Govt deficit. Haven’t you heard this identity?”
Vimothy – that’s because there was no real/nominal qualifier.
Wh10,
Cullen’s article is definitely timely. It seems to me that at times he’s making a very similar argument to the ones that I’ve just described, and at other times he’s making very reasonable statements, with which I can agree.
Here’s an example of the former:
“What this relationship [i.e. government deficit = private surplus] tells us is simply that as the government spends more the non-government’s income increases. This is an accounting identity.”
And an example of the latter:
“If the government turns on the spending spigot there is no guarantee that the private sector will continue to create goods and services that are of high quality that will increase our overall living standards.”
Vimothy,
As I read your arguments, by basic understanding is it does not matter much what the government spends or promotion of policy if it does not maintain or increase purchasing power at the individual agent level of the economy. i do not think MR would disagree with this if I framed your argument correctly.
Oilfield Trash,
One argument is over an identity like this:
Nominal private saving – nominal investment = nominal govt deficit + nominal current account balance
Ramanan thinks that the government can affect the *real* flow of saving simply by exploiting this accounting tautology. The government can make the difference between nominal private saving and nominal investment large, simply by making its nominal deficit large. This will raise private wealth and income, which are *real* variables.
But there is no *economic significance* to the difference between the two nominal flows. Whether the difference is very large or very small doesn’t mean anything, in and of itself. It’s just an arbitrary number. You cannot get the behaviour of real variables from a simple identity like this. You need actual *economic theory*. What Ramanan is doing here is accounting, which is something else.
“Ramanan thinks that the government can affect the *real* flow of saving simply by exploiting this accounting tautology. ”
You misrepresent me Vimothy.
You gave a situation where the deficit keeps increasing forever.
I said in the context of a closed economy it is highly unlikely because it will have the effect of raising private sector income and wealth and will as a result lead to higher private expenditure making the hypothetical situation very hypothetical.
Saying the government can exploit the tautology is different from saying the government can exploit its powers to have an effect on the economy. These two are two different things.
” Again, the causality is backwards. Since any outcome other than the fiscal stance and GDP growing at the same rate is unsustainable, you will not see it happen, as rule.”
If they were growing at the same rate AT THE SAME TIME you’d have a point The interesting part is Godley’s observation that AFS is a leading indicator of what nominal GDP will be in the future.
“So, for this period at least, the nominal GDP should track the AFS one for one, with a mean lag of only half a year. This is the amount of time that must elapse before the inflows become outflows, and it is well within the time span that would make the AFS relevant for policy-making purposes.”
Beowulf,
If they were growing at the same rate AT THE SAME TIME you’d have a point The interesting part is Godley’s observation that AFS is a leading indicator of what nominal GDP will be in the future.
That’s one interesting part, but I don’t see how it affects the argument about causality. However big the lag, if the fiscal stance and income are growing at different rates, sooner or later their paths are going to diverge, which is unsustainable.
Well yeah if I could write economic theory in one comment which every neoclassical economist could understand, the world wouldnt be in a problem!
Anyway I am going to try because you have already made up your mind, as far as I can see from your comments here and before.
Again, you keep talking extremes. Still!
It has taken me a lot of effort to set you straight on accounting. But I read carefully your comment and you still deny the identity.
(“Doesn’t actually increase its wealth”) you say.
That kind of result is valid only if the effect of *any* deficit is to neutralize the effect on nominal increase in wealth by reducing the real wealth by erosion via inflation.
But your argument isn’t restricted to that. You try to come up with your own invented accounting where somehow it is not even true at the nominal level because of vague future discounting.
If you so adamant about this point, I am not sure how to proceed.
It is hard to convince. You have in mind an exogenous supply side. Go and read some survey of producers and find out why they are producing more! These are the things Keynes tried to debunk years ago.
Oops meant not going to try!
Ramanan,
You’re either not reading my comments, or not understanding them.
I am not disputing the identity, which is trivially true. I am disputing the line of argument that goes: identity -> causality.
It has taken me a lot of effort to set you straight on accounting. But I read carefully your comment and you still deny the identity.
What do mean, “a lot of effort to set you straight on the accounting?” Please quote.
We are not arguing about accounting. If it was simply a matter of accounting, then the government could set the level of wealth wherever it likes, because NFA = NFL, or whatever.
Can the government set the level of private wealth wherever it likes?
I am still waiting to hear an *economic* argument for why, if the government increases the marginal quantity of bonds, potential output in the future rises. Why don’t you make one? I can accept that I might be wrong about this.
(“Doesn’t actually increase its wealth”) you say.
What I wrote was:
“More marginal bonds for the private sector doesn’t actually increase its wealth as a whole unless it increases its future income as a whole.”
If the government issues a bunch of bonds, then this might represent real wealth for the private sector, or it might not. It’s impossible to say without placing some more restrictions on our mental model of what’s going on.
The government can issue as many nominal liabilities as it likes. In nominal terms, the government make the private sector incredibly wealthy. Woo-hoo! But this is more or less meaningless, a trivial power that benefits no one. What it can’t do is make the private sector wealthy in real terms, simply by exploiting an accounting tautology.
Now, this does not mean that the government cannot increase private wealth by borrowing. It’s easy to imagine ways in which it can. For example, imagine that the economy is in some kind of bad equilibrium, away from its long run trend–like now–and that the government can borrow, pull in underutilised capacity and return output to its previous path. (I gave other examples upthread).
Then, the new claims on output (i.e. the extra bonds) would coincide with there being more output to claim. The government would have raised private wealth, in a real and not simply fictive, nominal sense, by borrowing, because the borrowing was tied to an activity that had a tangible impact on the economy’s ability to produce output.
But your argument isn’t restricted to that. You try to come up with your own invented accounting where somehow it is not even true at the nominal level because of vague future discounting.
Where have I “invented accounting?” Quote please.
What we are arguing about is not accounting but whether the existence of an identity that says that delta_NFA = delta_NFL means that the government can increase private wealth–not what the identity says or whether it exists, but what it’s *economic significance* is.
“I promise to pay you, Ramanan, $1 million dollars” (said in best Dr Evil voice). Feel any wealthier? If not, why not?