The imposition of a tax raises the price of things being taxed, and thereby generally reduces the amounts sold. This reduction in the amounts sold, relative to the situation before the things were taxed, is called the efficiency cost of the tax. This efficiency cost of the tax is usually shared between the sellers and buyers, in inverse proportion to market power of the sellers relative to the buyers.
The efficiency cost of the tax is best thought of as the sales that would have taken place, but for the imposition of the tax. These ‘lost sales’ represent things the seller could have produced (and therefore sold) but didn’t because of the imposition of the tax raised the price of the things. Therefore the seller couldn’t sell some of these things because of the price increased by the tax. In other words, the tax reduced the sellers production possibilities.
These ‘lost sales’ also represent things the buyers would have bought but didn’t because of the imposition of the tax raised the price of the things. In other words, the tax reduced the buyers consumption possibilities.
To summarise, imposing a tax raises the price of the things being taxed. And, as a consequence of the tax increase in price, both sellers and buyers are made worse off to the extent of the ‘lost sales’ – the so called ‘efficiency cost’ of the tax.
Consider income tax — it is obvious that firms would offer more jobs if labour was untaxed. In this case, the cost of labour would be lower than otherwise, thereby lowering the firms’ prices. Similarly, workers would be prepared to work more hours if they didn’t have to pay income tax because they would get to keep all of the additional income the extra hours generated. So, income tax reduces the amount of things produced as well as reducing the number of hours worked.
This efficiency cost of a tax (expressed in dollars and cents) is usually stated relative to the last dollar of tax revenue raised by that tax. For example, let us assume a very small change in an income rate produced one extra dollar in tax revenue. Also, assume that the efficiency cost of that change in income tax resulted in a loss of, say, thirty cents. Then you can say, broadly, that it costs the economy thirty cents to raise one dollar of income tax.
This result leads directly to the following implication for government spending. Governments should only spend tax revenue on things that provide a benefit greater than the sum of the tax revenue itself, and the associated efficiency costs of raising that revenue. (As an aside, technically, one should include the costs of collecting and enforcing the tax collection system. But we will ignore those for simplicity sake.)
To demonstrate the point, we will consider the example of the Commonwealth’s ‘investment’ in the NBN, using the numbers from recent articles by Kohler and Gottleibsen, and our example of 30 cent efficiency cost of one dollar of tax. By the way, I am not arguing that this scenario is correct. I am just using the scenario to demonstrate how the efficiency costs of taxation impact on the well being of the economy.
The original Commonwealth setup spending was of the order of $30 bilion. Recently, because the NBN couldn’t borrow from commercial banks the $20 billion it says it needs to fully roll out the network ,the Commonwealth has agreed to provide that as well as the initial $30 billion. However, according to Kohler, when it comes time to sell the Commonwealth’s equity in the NBN, commercial buyers will only pay $30 billion. This assumes that the NBN develops in accordance with its own corporate planning.
So the Commonweath has spent $50 billion but in this scenario but will only receive $30 billion from the eventual sale of the NBN. Now in addition to this loss of $20 billion, we also need to add the efficiency costs of raising that $20 billion. In this example, it is $6 billion (20 x 0.3 = 6). Accordingly, the all-up cost to the Commonwealth from the decisions associated with the NBN are likely to exceed $26 billion (20 + 6 = 26). From the Commonwealth’s perspective, not only has the NBN cost $20 billion in tax revenue but also the economy has suffered the ‘efficiency loss’ of $6 billion associated with raising the $20 billion in tax. And the Commonwealth has no income producing asset justifying such huge expenditures.
To put that amount in context, it is reported that the submarines to be constructed in Adelaide will cost in the vicinity of $50 billion. Hopefully the Commonwealth will end up with some operational submarines for that expenditure.
In finishing, I just wanted to point out another fairly obvious implication arising from the efficiency cost of taxation. In the context of fiscal consolidation, there are two polar opposites available to reduce a budget deficit. One is raising taxes, while the other is cutting spending. Here I just want to compare the different impacts of these opposites, but I recognise that both would likely be used.
Raising taxes can reduce the budget deficit, if the policy is correctly designed and the costings of the policy change are also correct. However, these additional taxes will cause efficiency losses in the economy making the economy poorer by these extra efficiency losses.
Cutting spending can also reduce the budget deficit, again assuming the change in spending is correctly designed and the costings of the policy change are correct as well. Provided the spending that is cut is of low quality (meaning that the spending doesn’t provide much that is highly valued), closing the budget deficit this way is superior to raising taxes because there is no or little efficiency cost, unlike with raising taxes.
I recognise that these efficiency issues have not been considered against the politics of either raising taxes or cutting spending. I leave those issues for others to consider.
Finally, given the efficiency costs of taxation, it is hard to agree with the Keyenesian approach of attempting to ‘create jobs’ by spending on anything without regard to the value purchased by such spending. Spending on Pink Batts and School Halls caused the ‘efficiency losses’ associated with the raising of the tax revenue to actually fund the spending. In addition, it also actually destroyed value in the economy because, I argue, that those that received the benefits of that spending, valued it less than the purchase price. If that wasn’t the case, they would have made those purchases themselves. In addition to that waste (the difference in the amount spent and the value to those that received the spending), you must add the efficiency cost of the taxes raised to fund the spending.