So in National’s tax package they have dumped the R&D tax credit. Businesses seem disappointed – but this isn’t enough information in of itself to tell me whether getting rid of the credit is socially optimal or not.
In order to analyse this we have to think about why an R&D tax credit could be socially optimal and then see whether that scheme actually worked in the appropriate way to increase social welfare.
Let’s give this a go 😉
R&D increases “productivity” so it is good right
This is a popular argument, however it is also one of the most misunderstood.
If R&D expenditure increases “productivity” then the whole benefit of this is captured within the market for the good. If the increase in productivity reduces the marginal cost of production, then the firm will increase production until the marginal benefit equals the marginal cost.
Now, the firm will want to invest in R&D until the benefit it makes from its investment is equal to the cost. As a result, if the firm extracts the WHOLE benefit of R&D it will make the socially optimal choice.
Since firms do not extract the whole benefit of production (as consumer surplus also increases in output) we may try to make the case that there is an externality here – however, this is likely to be inappropriate, as Coase bargaining is avaliable.
In this case, the price will adjust to take into account that there is some surplus associated with the R&D process which isn’t captured in the standard competitive process. This implies that the price will increase such that the market surplus gained from further R&D investment is split between the consumer and producer.
This makes sense as the transaction costs in this case are minimal – we are talking about an externality that exists intra-market, as a result there is a price mechanism which already exists to take account of it.
As a result, I wouldn’t take this as a reason for the R&D credit.
This is the kicker – there are positive externalities from some R&D investment.
The positive externality stems from “spillovers” in knowledge between firms. If one firm invests in R&D then this creates knowledge and ideas that also benefit other firms. As the individual firm does not take into account the benefit that there research has on other firms we end up with “less” R&D investment than is socially optimal (Note that I am assuming that efficient Coase bargaining solution will not appear – as I think there are sufficient transaction costs to prevent a co-operative equilibrium).
In this case we could use a tax credit to increase the amount of R&D investment towards the socially optimal level – as long as the benefit of this exceeded the cost of raising the funds this would be socially optimal.
However, this isn’t what the scheme was focused on it was looking at all R&D. Given the poor focus of the package I wouldn’t take this as a reason for the R&D credit.
Another area where R&D investment can increase social outcomes stems from the competitive environment.
Expenditure on R&D investment is “sunk” when the firm chooses its output level (competes) as a result, the cost of R&D does not impact on its output choice. However, higher levels of R&D reduce the cost (or increase the revenue) of producing an additional unit.
This reduction in marginal costs (or increase in marginal revenue) allows the firm to be sneaky and “commit” to producing a higher level of output when competition occurs. If a firm commits it can increase its profit – to the cost of the other firm. However, if both firms commit, we end up with higher output and lower profits – a situation which brings us closer to the competitive outcome, and as a result leads to greater social welfare.
By reducing the cost of this commitment resource, the government can increase the amount of “commitment” private firms invest in through R&D, pushing us towards the competitive outcome.
Now any gains in this sense will be incredibly small – especially given that the dynamic nature of the competitive process makes it likely that firms will solve the above “prisoner’s dilemma“. As a result, I wouldn’t take this as a reason for the R&D credit.
So there are multiple channel through which an R&D tax credit can increase social welfare. However, I haven’t found any of the channels particularly convincing. As a result, I’m going to stick to my prior belief that the tax credit was a dumb idea to start with.
People believe that things that “increase productivity” are good – however, this view ignores where the benefits of productivity go, and whether there is already a market mechanism that deals with that productivity.
If it was possible to observe the cases where positive spillovers exist then the government may be about to create a tax credit that would increase social outcomes – however, I am not convinced that we are in that case.
Furthermore, if you are wondering why businesses are complaining about missing out on it, think of it this way: The businesses that were going to invest in R&D anyway would be getting money from taxpayers to do it – of course it is in their interest to say it is a good idea!