I have a really basic question and I don't have time to go to the economics textbooks to check details. So I'm hoping someone can please shed some light on this.
Basically the story goes...
If the government spends X , then there will be m*X increase in GDP. (Fiscal multiplier effect)
I assume the number "m>1" is the multiplier that is estimated empirically and takes into account crowding out, etc. (Multiplier is because of money changing hands more than once)
When I learned this multiplier it was generally assumed that this somehow made government spending more preferable.
Isn't it the case that if an individual spends X the same thing will happen? Or is there some magic that happens when you collect taxes? Why does it matter that the government spend the money? I don't see what the trick is, is it so easy to debunk this? Please let me know if you have a reference to a discussion of this.
And most of all, what happened to assessing things based on opportunity cost, rather than some gdp accounting?! (rhetorical question)
All that matters within the Keynesian model is that money is spent, not who spends it. So you're perfectly right that individuals spending money should have the same effect. C+I+G+(E-M)=Y. It doesn't matter where the spending comes from, it just has to come from somewhere. This is why (contra many conservative and more than one libertarian strawman) oftentimes Keynesians propose a variety of tax-breaks. Indeed that was a large part of the 09 stimulus package.
The reason why government spending is supposed to be especially effective is that the government has an incentive to have no propensity to save and therefore it can activate the multiplier. This is why Keynes advocated the "treasure chest" idea of the government saving a huge amount of money during the boom period and then when the depression hits it uses this money in order to boost back up aggregate demand.
Keynesians say that since the tax cut multiplier b/(1-b), is one less than the government spending "multiplier" 1/(1-b)
(1/(1-b)) - (b/(1-b)) = (1-b)/(1-b) = 1
that "proves" that government spending is better for the economy than tax cuts.
That comes from disguising the "saved" fraction of disposable income (1-b)Yd as a+I+NX, and disguising tax T, as G.
It's another kind of Keynesian math fraud. I have more about it in Pt 2, Fiscal Multiplier Debunked.
Scroll down about 2/3 of the way to The Bogus Tax Cut “Multiplier” (TCM)
I answered you. Maybe I am not being clear.
Let's say GDP is 50, Consumption is 20, Investment is 30, Net Exports is 0, and Government Spending is 0.
Y = C + I + NX + G
50 = 20 + 30 + 0 + 0.
The government decides to raise $10 in revenue with taxation and immediately spends it. Now the equation will be:
50 = 15 + 25 + 0 + 10.
Consumption and Investment may not necessarily be those values but hopefully the point is clear.
There are several ways of decomposing GDP as the sum of a few accounting related aggregators.
The most common being Y=C+I+G+(X-M).
The rationale is that given an amount of goods produced during a year, they are either consumed by the private sector "C", saved as capital "I", consumed by government "G", exported "X". The imports figure "-M" enters to correct "C", "I" and "G", since some goods that are consumed or invested domestically were produced abroad.
And this is perfectly correct when applied to every specific good, and when the units are physical quantities, and not money.
The quantities of bananas produced in Republic of Banania is equal to the number of bananas eaten by the non-bureaucrats, plus the increase (or minus the decrease) of the aggregate stock of bananas, plus the bananas used up by bureaucrats, plus the bananas shipped abroad, minus the bananas we have counted before that were actually sent from other places.
Of course this holds logically true since any banana unit produced domestically is either consumed (i.e. destroyed), stocked or shipped away.
And if we don't know the total figure produced, we can still estimate it if we have a good idea of how many bananas fall on each of the above categories.
And what if we don't have total unit numbers, but we have instead payment receipts for banana transactions? Can we still estimate production?
Theoretically yes, but only when very unrealistic assumptions hold.
First of all prices of bananas units change over time and across the land. Of course we could keep track of prices of units for each payment receipt as well as the total amount, but that's the same as keeping track of units transacted.
And even if we assume that banana nominal prices are reasonably stable, we cannot estimate how many bananas were privately produced and consumed without being sold, or how many were privately saved under the mattress and not put in a deposit account. Since there are no transactions to any of these operations there are no payment receipts.
But even if we assume that these instances of banana autarkies are rare and negligible, there's another problem.
If we don't know the current level of the banana capital stock, we cannot determine how many bananas got in or out of it, and therefore we cannot uncouple what's actual production from what's capital consumption.
The Keynesian multiplier voodoo takes place due to these violations of accounting assumptions, specially the last one.
If some capital buffers (of bananas) could be somehow leaked and the proceedings of this operation sold in the market for domestic consumption or exported, there would be increases in "C", "E" or "X", without detectable changes in "I".
Of course, the real "I" would have to drop due to the consumption of pre-existing bananas, but since this was not accounted for it looks in the paper just like as if new bananas had been produced out of nothing.
The multiplier is thus explained.
But if the scheme goes on for too long, these leaked buffers of bananas will eventually be depleted, or their owners will notice the leak and move the remaining bananas to somewhere else safer.
Of course, all this applies to every tradable economic resource, not only to bananas.
You can even lump sum many items together, since figures are now in money terms, and not in good-specific units. And the result of this is the familiar GDP.
However one important item cannot enter in the above computation, and that is cash itself. More precisely, money liquidity. It cannot enter in the computation since it cannot be measured in money terms.
A GDP like indicator makes sense insofar as the "propensities to hold cash" remains the same. When they don't GDP returns will have hidden biases. That is, if liquidity increases, GDP returns will rise, and if it decreases, GDP returns will drop. However this does not reflect any effect on production, but only the non accounted oscillations on the liquidity profile. If we admit that people hold cash for rational reasons, the liquidity profile is indeed capital, and any shor-term increases on GDP figures are probably going to be costly in the future due to the distortions on the market liquidity profile.
So if government can leak capital buffers he can expend more without affecting directly consumption rates, as it would be the case through direct taxations. And as a bonus, the whole operation also boosts the accounted items in the GDP, at least in the short run. Since government performance perceptions and approval ratings are positively correlated with GDP returns, this is indeed very good news. And if the missing assets on the leaked capital buffers are only discovered during the next administration, that's a total jackpot.
What Keynes and his school discovered is that there are several ways government can perform these accounting schemes and effectively leak capital buffers, specially the highly "weird" and intangible liquidity-money buffer, uppon wich it share privileged access with the banking cartel. The basic scheme structure was already in place decades before Lord Keynes and his followers came along, but they provided intelectual backing and some new technical insights on procedural tactics.
Keynesians and the like usually seek to justify these schemes not on the grounds of the net benefits for governments and associated operations, but because they assume that sometimes, people's collective behavior towards hoarding liquid assets is "irrational", and thus create what they call liquidity traps.
That is, given certain economic injunctions, too many people would hold their money for more time, and given a fixed total amount of it, this would create shortages of liquidity that would self-intensify and create all sorts of economic disasters due to many businesses being unprepared to survive that.
So by leaking these private buffers of money through credit inflation plus government and/or private expenditures the liquidity returns to the economy and doomsday is postponed.
There are other "behavioral" reasons some keynesians invoke other than the liquidity trap. One that comes from Keynes itself is the notion of stickyness of certain prices and in particular the common belief that nominal salaries can not be lowered, so in order to mitigate unemployement, real salaries must go down with inflation.
That's all at least very debatable. And since we cannot run perfect parallel experiments a lot or room is left for speculation. The whole thing is undeniably very ingenious and mischievous. Of course this pattern can only be repeated a few times until the people that are being screwed figure out clever ways to protect themselves from it, but that's the fate of every such scheme.
In the end, to debunk some recommendation we should prove that it doesn't work as expected.The problem is to specify what are the expectations concerning the keynesian multipliers. Their effectiveness will depend on whether you consider them mere accounting schemes that are politically and economically profitable for those particular groups applying them, or whether you believe there are collective patterns of irrationality that emerge and can only be mitigated by this sort of large scale system intervention. For the first use, I'm sure they work, at least for a while until people figure them out. For the second use, I'm not convinced.
The large scale liquidity trap scenario for instance seems to be a feature of the single currency fiat monetary systems. Due the monopolistic powers granted them, the fractional reserve facilities (that is, the wholesale liquidity traders) all have incentives to be as much leveraged as they are allowed by the regulators. So when something unexpected happen and a more than a few payments fail, liquidity is quickly sucked back to fill the holes, but it's not enough and one large liquidity trader fails, and as it fail payments to various others in the system, they too fail in contagion and the whole system breaks down together. Since everyone is by law linked to the currency, there are few alternatives to avoid the hit and the costs of the whole mess are highly external. But since the benefits of leverage are internal, we have the classical situation of tragedy of commons.
In a multiple currency competitive setting, fractional reserve facilities would have more incentives to keep their leverage level under control, because they can no longer effectively externalize the costs of an eventual liquidity trap scenario if it is restricted to their own currency. Leverage levels would be thus correctly adjusted to market estimate of the liquidity risks of each currency.