National Australia Bank’s Business Confidence Index jumped from 3 to 7 points this week, a nine-month high coming on the back of this year’s federal budget. The ANZ-Roy-Morgan Consumer Confidence Index also rose after the budget.
This all sounds like good news but it also raises a number of questions. How good is the news? How long is the boost in confidence likely to last? And what might be different this year given more general economic conditions?
The first thing to note is that these measures track changes in business and consumer expectations. So if everyone knew perfectly what was going to be in the budget then, all else equal, it should have no effect on business or consumer confidence indices.
The second thing to note is that there is a lot going on in the background. The Reserve Bank is often tinkering with interest rates, the economic conditions of trading partners are changing, bond market conditions may be moving around. All sorts of things are going on. And while it is a fool’s errand to try and analytically control for these things – after all, you can’t control for everything – it’s worth remembering that they are going on.
A very dangerous and foolish thing for a Treasurer to do would be to attribute positive changes in business or consumer confidence right after the budget to that budget. That would be a classic illustration of the post hoc ergo propter hoc fallacy (from the Latin, meaning “after this, therefore because of this”).
Living in the moment
Lastly, these indices ask people how confident they are now. Ideally that could be interpreted as their best forecast of future economic conditions for the infinite future. But, as has been well documented by social psychologists and behavioural economists, people don’t typically act like this. Almost all of us – even those of us who know about it – suffer from some sort of “present bias”. We put more weight on today versus the future than we really should.
Last year’s budget was widely considered to be a political disaster – one from which the government is still trying to recover – yet it’s correlation with business confidence was hard to see. The NAB Business Confidence index went from 7 to 7 after last year’s budget, and the in the following two months went to 8 and then 10.
All this against the backdrop of it seeming fairly likely that many budget measures designed to reduce the deficit were going to be blocked by the ALP and that the crossbench were hostile. It did drop significantly in September and then further, but it’s unclear that there was a big update about the likelihood of the measures passing by that stage.
Now, correlation isn’t causation, but it would seem hard to make the case that last year’s poorly received budget had a causal impact on business confidence. Interestingly, consumer confidence did fall, perhaps indicating a downward revision in household expectations about disposable incomes.
One conjecture is that budgets have much more of an impact on consumer confidence than they do on business confidence. Perhaps business confidence is driven much more by general economic conditions such as: interest rates, credit availability, and global economic conditions.
But for households, the budget often has a real impact on disposable incomes. Tax and benefit changes have a direct impact on consumers. Moreover, these are often quite a surprise (think pension indexation, the family tax benefit, and the 2% deficit levy from 2014). And as mentioned above, it is surprises in expectations, not the general level of expectations, that affect confidence indexes.
Looking at the historical data, the indices move around a fair amount in the coming months. Perhaps this reflects present bias among the respondents.
In light of all of this I, myself, would not put a lot of stock in how business or consumer confidence changes after a federal budget is handed down. Maybe changes in consumer confidence give some idea of how households are going to change their expenditures, and that has an impact on the macro economy.
But if I wanted a quick look at how the budget was received I’d look at revealed preference measures of behaviour. Are business investing in capital and hiring? Not according to the latest capital expenditure figures. Are consumers spending on durable goods?
These are not perfect measures, but arguably they are better.