Parental Support: The Elusive Market Failure

Yesterday I argued that when it comes to parental support what people seem to want is a frictionless ability to transition between work and home lives. Specifically, the ability to give birth to a child, take care of that child away from work for a period of time and then to return to work with the same prospects and opportunities as the day they left. That said, there is the issue of whether it is feasible to return to work in a frictionless way given possible depreciation in human capital and also whether there has not been a change in the way parents want to work when they have children (e.g., they may not want or it may not be socially desirable to want them to put in the same level of hours or not).

Today, in order to evaluate policy rationales for parental support, I do the usual economic thing and ask whether the market, left to itself, will get it right or not. Of course, that leads to a natural question: what does getting it right look like?

It turns out that the answer to that is not too difficult. When parents spend time with the children they are forgoing other activities — let’s focus on work. That involves a cost both in terms of what value parents get from work (and by this I mean work satisfaction and not money) and also what employers get from those workers. This latter one is tricky as it depends upon the nature of the labour market for the parents’ jobs. If there are many substitute workers available, employers can just replace the workers and so incur no cost. On the other hand, if substitutes are either not available or there are costs in employing them for relatively short periods of time (say, because of firm-specific training required), then employers face real costs.

What of the benefits? First, there is the raft of benefits that parents get from spending time with their children. And to the extent that there are developmental benefits to the child, the parents feel that benefit too. (There are costs there too but let’s assume that, for the most part, there are positive net benefits). Second, it may be that, like the benefits of education, the benefits associated with child development also spillover to the rest of society. That is, there could be positive externalities that are not captured by the parents. Note, however, that if you argue this then you are arguing that parents would not, of their own accord, make decisions on child development that would be sufficient for society. This is a notion that many people appear to be uncomfortable with.

But there are caveats to these benefits. Like cost, you have to ask, benefits relative to what? After all, many of the child development benefits can be achieved without specific parental involvement. Indeed, parental involvement is likely not necessary all of the time. Non-parent child carers can do the job and in some cases, let’s face it, may do things better. Even with breastfeeding there are technological options that reduce the need for 24/7 maternal availability and those only make the private ability to engage in breastfeeding easier (or at least no harder).

The market will operate well if whenever those benefits exceed those costs, parents are not only able, but choose, to take parental leave. Otherwise, there will be a market failure and a potential rationale for policy intervention.

Now I am not going to be able to prove a particular market failure exists or not here. However, I think there are some candidates that we can list. First, and foremost, is the problem of indivisibility. This is the idea that some jobs cannot easily be broken up into smaller bits whether it be on a day by day, week by week or career basis. It is career indivisibility that presents a problem because it says that workers who are away from their jobs for long periods or who cannot commit to long work weeks may be less productive than those who do. Productivity includes here working with teams, the demands of a firm’s customers (e.g., this is particularly salient for legal partnerships) and the potential depreciation of training and job-specific skills. That said, it is not a given that jobs are not scalable. Indeed, the ability to have a family life may be quite compatible with job performance and may enhance it. My point is that there are some jobs that do not have this characteristic.

What this means is that, in a competitive labour market (or a competitive product market), workers who cannot commit to their work may not be able to access those jobs that require such commitment. If there are many jobs of this kind, then there may be an under-supply of workers to such jobs (the so-called skill shortages) and also, for those who do undertake them, on-going pressures to over-supply (the so-called rat race).

That said: indivisibility may just be a fact of economic life, where is the inefficiency? My guess is that it is this type of pressure that forces intra-household specialisation — that is, one parent goes to work while the other does not. To the extent, that single parent involvement is not best for the children, then this might give rise to other effects.

And the ultimate culprit here might be that there is too much competition in the labour market. Now I don’t want to say this lightly but this is a ‘thought piece.’ Think about why we have mandated annual leave requirements. You can say that firms would surely find giving workers some time off a productive thing. But it may also be that in a competitive market-place they find it hard to commit to it. The mandated requirement breaks the indivisibility by restricting the labour that everyone can supply. It potentially makes a larger group of people better off who could not otherwise express that preference in the economy. The same considerations may apply for parental support.

Ultimately, it is indivisibility that is potentially driving other side-effects with more clear social consequences. So let’s move on to those.

Second, there may be externalities associated with child development. Stuffing up childhood doesn’t make for socially and economically productive citizens in the future. But to the extent that work and income pressures lead to some households being unable to supply the requisite level of child development to stop this from happening, we have a problem. I suspect, however, that this is mainly a problem for low income households. To be sure, higher income parents may neglect their children. However, it is not because they do not have an economic position to choose not to do that. It is likely other things. Lower income households may not have that choice and the children might suffer from the impact of temporary poverty. It is the potential externality associated with this that can motivate a government role.

Third, there is the potential for discrimination; not just on the basis of gender but on the basis of any preference to spend more time with one’s children. Consider an employer who has one of those jobs that requires full attention. Suppose also that the employer has to invest in the worker for that job. If you were that employer and you faced candidates from identifiable groups (say, a man versus a woman) who have differing statistical likelihoods of wanting to scale back at some point who are you going to take a bet on? And if every employer decides the same way, what does that do to the statistics? The answer: it reaffirms them and you have gender roles and a pay differential between genders. And the inefficiency here is that the wrong people may be ending up in the wrong jobs.

And this discrimination can take many forms. The fact that when a baby arrives, maternity leave is taken much much much more than paternity leave may be because of stereotyping that constrain fathers who would like the reverse to be true. It is not an unreasonable hypothesis that one of reasons that fathers are not taking time off is that the discrimination against them is far worse than against mothers. They face greater stigma, far greater harm from signaling their family-oriented preferences relate to norms and these add up to greater costs associated with staying at home.

Finally, there are imperfections in financial markets that impact upon life-time household decision-making. One of the ways a household could manage all of this would be to be able to hit the career pause button but be able to smooth the consumption consequences of this (most notably on housing payments) over their life-time. But how many banks do you know that allow mortgage repayments to stop following the birth of a child? And why should they? They don’t know whether the pause is short or long? Consequently, many families who have the potential life-time income to support a parental pause, cannot do so because of liquidity constraints.

So there are candidates for market failure here. However, unlike other forms of market failure, they are not straightforward and impact on different people very differently. There is also limited evidence on their magnitude or relevance. All this adds up to a difficult time assessing potential policy responses and a mess of unintended consequences associated with many of these. But I’ll wait until tomorrow to drift into that quagmire.

12 thoughts on “Parental Support: The Elusive Market Failure”

  1. What about if children just like having their parents around? (Having just had a kid myself, I think they do) There doesn’t have to be any long term benefit, but there is an externality there that the parent doesn’t internalize. Even if the child’s utility enters the parents’ utility function, parental care will still lower than the efficient level. Good luck getting a number on this though!


  2. Actually, after giving 2 seconds more thought, if we assume that the child gets any benefit at all from being cared for, no matter who cares for it, and it can’t pay for the care itself (for obvious reasons) then there is an argument for subsidized child care.


  3. If the child gets utility and the parent internalises that utility, then there is no externality; someone has it covered and they are getting the benefit too.

    However, if they are not fully internalising the child’s utility then there is an externality but I put that under the entire rubik of child development.


  4. The issue with most markets is that money value is not the only consideration when making choices. The value of markets is that it allows people to make choices. When the main measure of choice is simply the current monetary value associated with some activity then it will dominate the reason for making the choice. How to overcome? Well put more information onto the measure of choice (namely money). The contingent loans that Andrew addressed in his AFR piece is one way. A person who takes parental leave continues to get paid at some rate (they choose) but “pays” back some of the lost time in the future because the money they receive during parental leave (perhaps money supplied by the government) will get paid back in the future. Instead of baby bonus we give children money – but it has conditions attached to it. It can only be spent on education – perhaps preschool, perhaps university, perhaps trade training but the parents and then the child get to choose the most appropriate use of their education money. In other words putting information and “rules” onto money gives a mechanism that allows all sorts of social and community policy to be simply, easily, and cheaply implemented. It also gives us a way of being able to measure the outcomes because we can examine the situation of people when they spend the money or repay it. If the rules are not working as expected then we change them – but we do it in the light of experience not by some computer modelling of likely outcomes.


  5. Josh, you talk about the externalities (I presume negative) from ‘stuffing up’ a child’s childhood – and thus implicitly the positive externalities are those that flow from not having a stuffed up childhood.

    But we are not talking here about years out of the workforce (if we are that is a very different thing). The Productivity Commission’s inquiry is specifically about parental support for newborn children – from what I can tell, looking only at government support in those first 12-14 weeks of a newborn’s life (might be up to the first 6 mnonths, not sure).

    So, what are the specific positive externalities from paid paretnal support in this case?


  6. Josh, I think you misunderstand, or I misunderstand you.
    The parent gets utility that is a function of their own consumption and the child’s utility. Following the obvious notation:
    They maximize this. However, total utility of the household unit is:
    Less u will be provided than optimal. Caring about the kid is different from internalizing their utility.


  7. I haven’t read Becker’s stuff on the family, so I’m not sure what the RKT is (great, another thing to add to the list!), but I stand by the claim the u will be below the optimal level in the above problem. Just maximize subject to a budget constraint, although I may be missing something here.
    I guess the relevant point is whether parents internalize or care about their kids in the sense I laid out above. But why would they internalize? What’s their incentive?


  8. Having now spent 2 minutes looking at the rotten kid theorem, I’m not sure how it would apply to this situation. You might be able to develop some sort of incentive structure here, but to my mind it would have to be quite contrived. However, my prior on this is very flat, so I think I could be convinced otherwise quite easily.


  9. Let me put it this way: I don’t think the externality on the child is going to motivate actual public policy in this space.

    That said, the RKT is a subtle and beautiful theorem that my hunch is could take care of this issue but I don’t have time to get into it.


  10. If you are saying this argument wouldn’t influence the politics, I agree. However, this externality argument could influence my opinion.


Comments are closed.