The economic impact of the coronavirus in New Zealand

In a recent post I discussed the impact on the broad economy associated with the coronavirus.  However, this is only a starting point for thinking about economic impacts – the next question is how we can understand the composition of the shocks, how we can measure this in real time, and how we can consider the areas where policy is relevant.

This is something I want to discuss here.

In that regard the Spin Off just had a good article talking about economic consequences, and interest.co.nz also had a good piece talking to the bank economists.  Finally, Westpac released a bulletin that discusses what they think is important. This is a complement to their pieces as I want to use the same “demand” and “supply” shock analysis as I did in the prior post to bring some of these concepts out.

What I’m discussing below is how I would look at this sort of crisis in real time as an interested observer – I work in research not policy, so I see this as a chance to open up a dialogue with other interested people in the comments below.  Any insights you have would be richly appreciated.

Furthermore, as I just don’t have the data on hand I would like (again I work in research, not as a forecaster, an investment analyst, or a policy person) I can only talk about what I would use – if anyone has been using this data and can discuss trends it would be great to chat about this in comments!

Note: Thanks to Matt Nolan for discussing this with me, and helping me to get the right data sources for this post.

Scenario 1: No virus in New Zealand – focus on the economic effects

As with the previous post it is best to have two separate scenarios, so let’s start with the first one first – when the virus is significant globally but doesn’t reach New Zealand.  This is a good place to start to think about this issue as it also represents where we are at the moment.  

In the previous post we noted this was a temporary negative demand shock.  So how can we think about the composition of the demand shock.

As a temporary demand shock there is not an initial need for the government to do anything – once the sickness has passed in a couple of months demand will return to normal.  Any short-term disruption is best dealt with by the central bank reacting, while individual businesses should be able to deal with what is a relatively small shock.

However, if the demand shock is longer but not permanent there is a concern that businesses that are exposed to the shock may be unable to insure themselves against this – with consequences for employment and knowledge in the economy as the firms disappear before the shock reverses.  Such a shock is temporary but persistent.

In other words, we are worried about business failure from a prolonged period of weak demand in specific sectors there could be a case for government insurance.  In this case we need to identify the sectors that are effected, and ask what form of assistance is appropriate. I would look to similar programs that already exist (eg payments to farmers during severe droughts) as a form of social insurance.

The composition of the demand shock

How can we think about the composition of the demand shock – this is an issue I’ve heard people discussing on the street, so it is certainly an issue of interest!  Furthermore if the shock is temporary but persistent there may be a policy motivation for helping out.

So what are the key areas where demand is expected to drop in this case?  We discussed this last time:

  1. Tourism (and export education),
  2. Export trade,
  3. Consumer and business confidence.

But what indicators – or bits of data – can we look at to try to understand what the costs could be? 

Tourism and export education

On the tourism numbers there is not too much so far, as the airport changes are recent. So the international travel numbers will be a bit late. [Side note: There will also be an interruption in terms of international migration – but this is likely to reverse out and so will be put to the side]

However for the urgent matters, the Ministry of Transport is the first contact to check if they have daily or weekly figures on “port” arrivals from air in New Zealand which could be analysed to try to understand what the initial shock is.  

We could then come up with a rough estimate of the drop in expenditure based on the decline in numbers, and the spending estimates from the tourism satellite account.  Using the daily data could provide a good counterpoint to the macro estimates from private forecasters.

Without those numbers another idea is to ask what happened with tourism and SARS – old NZ Herald articles point to a drop, but any other research on this could be useful. Also this drop needs to be corrected for the fact that tourist arrivals from China are now a much larger percentage of total tourism than they were in 2003 – implying the “shock” is larger!

For the export education figures – it would be good to chat with the big providers (eg universities and polytechs). Matt tells me he keeps receiving emails from the university about the interruptions associated with the border closures, so they have the data – and that the service Tertiary Insights (formerly Education Directions) is also providing good information.

Export trade

Here it is difficult to figure out how trade will drop – but if we have an idea of where trade occurs we can tell risks.  And we can work this out by looking at the Overseas Merchandise Trade indicies to understand the share of what and where New Zealand firms sell overseas.

One issue is that this doesn’t separate between the “volume” and the “price” (later quarterly statistics do) – and this difference matters for thinking about the economic consequences.

As a side note, if the virus constrains imports from China could also put up some pressure on import costs for both New Zealand firms and consumers, which might have more general negative consequences – although there is little policy can do in this space.

Staying on the export side we need to break down things a bit more.  Product by product is a very laborious task, instead we can think about “durable” and “non-durable” exports.

Let’s look at forestry as an example.  It is heavily influenced by what is happening, with trade to China stalling. But the logs can be stored as inventory as they are durable.  As a result, this is partially reversible.

With non-durable (meat, dairy) the produce has been created and will simply be destroyed if it is not sold – as a result, it needs to be sold in a market for a lower price or essentially lost.  The reduction in demand will lead to a drop in price which is a permanent loss in income.

In this way, it is likely to be providers of non-durable products that have the most to lose from the crisis – while durable sellers face issues of liquidity.

So how do we measure these shocks?

For non-durables it is all about price, so we need spot market estimates.  Commodity futures (eg on Bloomberg) will give a good idea here. For dairy we have the Fonterra GDT every fortnight which did fall sharply but I would expect a much larger fall in prices going forward.  Scenario analysis (looking at a range in price shocks) can help to think through the risks.

For the durable goods we want to understand quantity and price – so if there is any information on logging operations or mills shutting down that would be valuable.

Consumer and business confidence

Finally we have the unknown general demand impact – from consumer and business confidence.

Surveys of these measures are relatively infrequent (at least for a sudden crisis) so they can only take us so far.  Instead Google Trends can give us a great heads up on what is going on.

More broadly, without confidence numbers we can ask about history and other countries – what has happened to confidence in areas that have had recent surveys? What has happen with the stock market as an indicator of the priced in cost of the the crisis?

Using this to understand the consequences

Given these specific shocks to certain sectors, there are questions about risks to enterprise operation (will firms with knowledge shut down due to uninsurable risk?), employment (are there consequences in terms of jobs).  Furthermore, there is a question of what areas will face the most pressure (what parts of New Zealand are most exposed to these shocks). 

The way to look at this would be to construct an index of industries that are most exposed, and then look at GDP and employment shares in the exposed sectors – as a first approximation.  Given this information is available by region online this gives a quick and rough way of thinking where the risks are – and where policy might be usefully deployed.  I have seen this type of data produced in government such as here, which may be used for this.

Matt recommends a tool his old work has for this type of work, which can be found on the Infometrics site here. This can be used to easily construct a “sector” that we think is vulnerable and then quickly see its employment, GDP, and business count shares by region.  That sounds good, but one needs to pay a subscription for the service so I can’t really go through that process in this post 🙂

Scenario 2: When it comes here

For the second scenario we can keep the same shock and add an additional drop in retail demand and also hours of work based on the severity of the sickness in New Zealand (and additional confidence effects) – this requires more close up look and a general set of clear scenarios for scenario analysis.  

However, the key point is these shocks would be an addition to the more clearly defined sector shock posited above.

What becomes much more important for policy is identifying coordination failures.

The clearest example is with regards to how to stop or slow the spread of the virus, as a communicable disease implies there are externalities from an individual’s choice.

Let’s say employees and employers are averse to using sick leave – but not using it increases the chance of transmission which has an externality.  Do we need to advertise the importance of taking leave? At what point are schools and public offices closed? When are non-essential services closed to limit the spread of the virus?

These are public health issues, where what might be best for the individual has severe consequences for the other individuals around them – and so restrictions could be necessary.  But how to judge that?

No GDP figure, or appeal to demand and supply can inform us on that question – so when the virus reaches here the government’s main role is in making judgement calls about these trade-offs.  And the some of the best advice will come from epidemiologists and public health professionals regarding these specifics, not so much economists.