I am having a debate with somebody about VAT (sales tax) and whether an increase in the rate can cause price inflation. I am sure it cannot (price inflation is always and everywhere a monetary phenomenon), but I have confused myself and I’m not sure where I’m going wrong.
I understand why an increase in the price of a single good (like oil) cannot cause a general increase in prices. Assuming no change in supply or demand for money, the increased spending on oil and oil-based products will be offset by decreased spending on non-oil products such that the general ‘price level’ is unchanged.
The UK government has just increased VAT from 17.5% to 20%. So people will now be spending more on products on which VAT is charged, so they will have to spend less on non-VAT products. However, unlike in the previous case, the extra VAT paid is not “sunk” but will be spent by the government. In theory, the government could spend its extra revenue on precisely those products where individuals have now reduced spending. This would mean the price of non-VAT products would not change, but the price of VAT-products has increased, so the general price level will increase.
I’m sure this cannot be right. Can somebody tell me where I am going wrong?
Wouldn’t a VAT tax cause people spend more on non-VAT products? I think you are wrong in assuming that a tax won’t change people’s buying patterns.
Also, if the government spent all its money restoring the price levels, it wouldn’t have any left over to consolidate its own power. What would be the point of the whole exercise?
Only if the quantity of sales of VAT products falls. But then the prices of the VAT-products would not rise as much and the two effects would cancel each other out I think.
Certainly, it will change buying patterns, but I am assuming it won’t. I am making ceteris paribus assumptions to try and isolate the effects of the tax increase alone.
My question is purely an economic one. I agree it would be completely pointless for a government to do that.
Or, they will reduce the amount of such products they consume.
Yes, the government will “pick up the slack” in the fabric-softener industry… *rolling eyes. Back in the real world, the government uses its revenues to subsidize idleness (welfare) and police/military spending.
Yes, you’re trying to apply monetary analysis to supply and demand when you should be doing the opposite. Any sort of tax - regardless of its form - creates a “hysteresis” or differential between the price paid and the profits earned. If I tax a product at x% of “value-added” (whatever that’s supposed to mean), it means that there is less profit for the producer and a higher price for the consumer than would obtain in the absence of that tax. In the case of targeted taxes, this has well-understood effects, namely, it pushes consumer demand for the targeted item down the demand curve to the higher after-tax price-point and it pushes producer supply down the supply curve to the lower after-tax price point. This is mainstream economics, even Keynesians don’t dispute this (they just make magical fairy claims about how the unicorn taxes spent by Wonder Government cause the entire world to blossom into Olympian super-abundance). As a result of the tax, real production is below the point where supply and demand curves meet and the difference in production between the real and the ideal is termed “deadweight loss.” Since items with highly inelastic demand curves are less visibly affected by this, the government tends to target those items more heavily (mainly, vices… alcohol, cigarettes, strip clubs, etc.)
Secular price inflation (or deflation) is the result of changes in two variables - the supply of money and the demand for money. In a fiat monetary system, the money supply almost always swamps the demand for money so changes in the demand for money become negligible and we can talk about price inflation solely in terms of changes in the money supply. Otherwise, ordinary supply and demand analysis apply. If the demand for money goes up, prices go down (money becomes more valuable and gains purchasing power). If the demand for money goes down, prices go up. And vice-versa for the money supply. Taxes can only be said to cause inflation or deflation insomuch as they affect the supply of or demand for money. I am not aware of any scholarly economic works that make a case for a link between taxation and systematic changes in the supply of or demand for money.
I agree completely with your last paragraph. I am just trying to make sure I know exactly why “Taxes can only be said to cause inflation or deflation insomuch as they affect the supply of or demand for money.” I am sure you are right, but I have been challenged on it and I’m trying to defend this statement.
Perhaps a scenario will make my difficulty clearer. I will switch from VAT to a targetted tax on cigarettes, to make the scenario simpler. I will assume that cigarettes are perfectly inelastic, because I think to talk about people buying less quantity of goods because of the tax is confusing the issue. If taxes can truly never cause a general increase in prices, then it cannot do so even under the assumption of perfect inelasticity. Assume throughout that there is no change in the supply or demand for money.
Suppose cigarettes cost £5 now, and the government introduces a 100% tax on cigarettes. The price of cigarettes becomes £10 overnight. Assuming perfect inelasticity, everyone will be spending £5 more on cigarettes, so they will be spending £5 less on non-cigarettes. This will cause the general price level of non-cigarette goods to fall, but this would be exactly offset by the increase in the price of cigarettes, so there will be no price inflation or price deflation at this point.
But then the government spends all the £5’s it gets from the cigarette tax on guns. The price of guns will increase because of this extra demand. As the price of guns is included in the ‘average price level’, there will be price inflation.
This will cause the general price level of non-cigarette goods to fall, but this would be exactly offset by the increase in the price of cigarettes, so there will be no price inflation or price deflation at this point.
because you are using a ‘mainstream’ concept of inflation, this will depend entirely on how the inflation index is populated and weighted. what do the administrators of the price index want to show … ?
Let’s make another assumption. The £5 that people now spend on cigarettes was previously spent, by everybody, on milk. That is, everyone has decided to cut back on milk in response to having to pay more for cigarettes. Now only three goods are affected: cigarettes, guns and milk. Cigarettes and guns will increase in price, milk will decrease. What will happen to the price index of cigarettes, guns and milk?
I guess it all depends how the index is weighted. Am I right in thinking that there is no such thing as an ‘unbiased’ weighting, because quantities of cigarettes, guns and milk are all in different units?
VAT is basically a production cost that will decrease the marginal value productivity of labor. Like other taxes, it will initially transfer wealth by income distribution (although indirectly so) from the productive to the non-productive. The total effect, of course, will be less productivity then there would otherwise be in the absent of such a pernicious tax. Less productivity means things are more expensive.
But when you ask about price inflation, you are basically asking about an increase in prices over time. There may or may not be such a visible increase due to the introduction of a VAT or any other tax because there are other forces at work that may still cause the general decline of prices. This is why the government will nearly always get away with such destructionist policies. The only thing that we can conclude is that, all other things being equal, prices are more expensive then they otherwise would have been if no VAT was introduced.
Totally understood. The Austrian way is to think about ceteris paribus effects, like we do when we think of changes in the world of the ERE.
Over time, yes. But can we assume this away as well; assume no change in productivity?
Only because of the reduced productivity, yes? So if we assume that away, all that we can say is that some prices have increased, others have decreased. Whether the ‘average price level’ has changed depends entirely on what is included in this measure and how it is weighted.
There is no necessary change to the demand for money (which, we all agree, would cause ‘price inflation’ or ‘a general increase in prices’), resulting from the imposition of a tax. But the decline in productivity that would likely result from the tax would cause the demand for money to decrease (“fewer goods chasing the same amount of money”), hence a general increase in prices.
"Over time, yes. But can we assume this away as well; assume no change in productivity? "
To assume that away would be to assume the cost of the tax away. And to assume that would be to assume that the tax is basically harmless or that it doesn’t exist in the first place.
The immediate cost of the tax, in this scenario, is that cigarette smokers have less milk. A further cost comes when the government spends the money, when it increases the price of guns. If the tax also results in decreased productivity, then this will be a further cost of the tax. But the ‘if’ condition is not a praxeological certainty, so I think it is still meaningful to analyze the effects of the tax without assuming that the tax will result in decreased productivity over time.
I know that the scenario is far, far from the real world, but I am trying to create a model in my mind that will enable me to see the effects of the tax in isolation. Kind of like how Mises uses the ERE to isolate profits from interest, and clarify other concepts.
it is a praxeological certainty; the tax reduces both the marginal productivity of labour and that of capital. by fiat, it declares all those previously profitable enterprises to be less profitable (less productive of phsycic profit to the agents)
“The immediate cost of the tax, in this scenario, is that cigarette smokers have less milk.”
A VAT is not a consumption tax. I’m not sure how you want to target a VAT only on a particular good.
“so I think it is still meaningful to analyze the effects of the tax without assuming that the tax will result in decreased productivity over time.”
As long as the government spends the money it taxes away, there is always a transfer of productive resources to non-productive resources. You can’t assume then, no curtailment of productivity.
It’s like this. There is a pool of productive resources you can tax. Once you tax and then spend, those resources that are diverted in order to satisfy the government spending are taken out of the productive pool of resources. They are no longer tax payers but tax consumers. Their income statements may say otherwise, but the economic reality is that they have contributed nothing to the pool of taxable funds. The pool must shrink as a result of any tax.
There is only a finite number of, say, British pounds in people’s wallets. We are assuming that that number stays the same, meaning the govt doesnt print more.
OK, to make it even simpler, say one person has 100 pounds in his wallet. Usually he buys a certain basket of goodies every week, which costs a hundred pounds. Today he goes to the store and finds out that prices of some things have changed, due to VAT. So the basket of goodies he buys may be different this week, because of the new situation. Maybe he won’t buy as much of item X as he used to, because buying the same amount will leave less money over for things he wants more than Item X.
Who knows what he will buy in the end? Not me. But one thing is absolutely certain. He won’t spend more than his usual 100 pounds, BECAUSE HE DOESN’T HAVE MORE THAN 100 POUNDS IN HIS WALLET in the first place.
“Ah but the govt will spend more.”
Well, let’s look at it this way. It’s the same thing as if there was no VAT, but there was a policeman at the entrance to the store, taking away 10 pounds from everyone who came in. So the guy will spend 90 pounds, the policeman will spend 10. Total 100 pounds.
As long as the number of pounds out there does not increase by money printing, it is just physically impossible to spend more than there is, right?
I see now that reduced productivity is a praxeological certainty following the introduction of the tax. To put it in terms of my scenario, the amount of milk being produced will necessarily decrease. This will lower the marginal productivity of land, labor and capital in general.
With less goods being produced, the demand for money must necessarily decrease, so prices must necessarily rise. [I think this is right, and is Say’s Law, essentially “supply constitutes demand”]
So actually, it IS a praxelogical certainty that a tax will cause a general increase in prices. It’s still true that (as Clayton said) a tax only causes price inflation to the extent that it changes the supply or demand for money, but the fact is that a tax will always (ceteris paribus) decrease the demand for money. So it is true that “the increase in VAT will cause price inflation”. There is just an extra step involved in the reasoning.
Whether it will cause the CPI or any particular index to increase depends on the constituents and weighting of the index. But if we use a more solid definition of “price inflation” meaning “a decrease in the price of money”, then it is true that the increase in VAT will cause price inflation.
DD5, I don’t know what you mean by “A VAT is not a consumption tax. I’m not sure how you want to target a VAT only on a particular good.” How is it not a consumption tax? And the reason I switched from VAT (applicable to some set of goods, but not all) to a tax on cigarettes is to simplify the scenario. They are equivalent, in that there is a set of goods being taxed, and a set of goods not being taxed.
Clayton, you said “I am not aware of any scholarly economic works that make a case for a link between taxation and systematic changes in the supply of or demand for money.” But doesn’t Say’s Law do this? I mean, the link is simply (1) taxation reduces productivity, so there are (2) less goods chasing the same amount of money, which is the same as saying (3) demand for money has decreased.
"So it is true that “the increase in VAT will cause price inflation”. There is just an extra step involved in the reasoning. "
No. Demand to hold money may rise, fall, or stay the same. The tax will, ceteris paribus, cause a rise in prices because the same amount of money is now chasing less goods. No requirement of change in the demand to hold money is required.
"DD5, I don’t know what you mean by "A VAT is not a consumption tax. "
What I mean is that it is not the typical consumption tax where you can just tax cigarettes or liquor. it is a tax on the value added by each firm and business at every stage of production. It would be practically impossible to only tax all those stages of production that have only to do with cigarettes.
Really? I thought that “same money chasing more/less goods” was like a metaphor for a change in the demand for money, alluding to Say’s Law. The demand for one good is bound by the supply of all other goods. (If there was only one good in the economy, there could not be any demand for it.) So less non-money goods means less demand for money, ceteris paribus.
Or to put it another way. We know the price of money is dependent on only two variables: supply and demand for money. You say the tax will, ceteris paribus, cause a rise in prices, which means the same thing as saying price of money has fallen. The supply hasn’t changed, so demand must have decreased. No?
Thanks, I did not know that, I thought it was only consumer products. My difficulty here is more general than VAT, since it applies to straightforward consumption taxes as well.