Monday, May 23, 2011

How well can volunteering be explained by a naive economic model?

I should begin by defining what I mean by a naïve economic model. The naïve model I have in mind is a conventional neoclassical model, with a few bells and whistles added. The bells and whistles are necessary because so called ‘rational economic man’ who is the basis of conventional neoclassical economics doesn’t practice altruism. There are probably still some economists who claim that everything everyone does is for a selfish reason, but I am not one of them. While I recognize that a lot of people do a lot of noble things for their own satisfaction, I see no reason to doubt people when they claim to be motivated by altruism.

So, in terms of the naïve model I have in mind, the objective functions that individuals follow in making choices take some account of the well-being of other people (i.e. I am assuming interdependent utilities). That means that individuals might volunteer to do something even if they perceive that this involves some cost to their own well-being. The extent that they do this would depend on the net cost in terms of loss of individual well-being and the extent that their actions affect the collective benefit they seek to obtain by volunteering. The main potential source of net loss of individual well-being would be the value to the individual of opportunities foregone from use of time in volunteering, which would be offset to the extent that the individual obtains satisfaction from volunteering, or from recognition of her efforts. The effect of individual actions on the collective benefit being sought would depend on the size of the group seeking the collective benefit. In a large groups the actions of each individual tend make a small contribution to the objective being sought, so there would be a greater incentive to free-ride on the efforts of others.

The naïve model suggests to me that people would tend to volunteer to a greater extent when they had fewer opportunities for paid employment. It therefore suggests that volunteering would tend to decline if workforce participation increased. It also suggests that volunteering would be a substitute for other forms of charitable giving – people with time on their hands would tend to volunteer their time and people in well-paying jobs that give them little leisure would be more inclined to put their hands in their pockets to make financial donations. It also suggests that people would tend to volunteer to a greater extent in small, well-defined communities (e.g. country towns) where their efforts are more likely to be recognized that in major urban centres where individuals are more likely to get lost in the crowd.

How well does this naïve model explain volunteering in Australia? Not particularly well. The first point I noticed when I looked at the relevant section of the Productivity Commission’s recent report on ‘Contribution of the Not-for-profit sector’, is that there has been a consistent upward trend in rates of volunteering across all age groups over the last decade, although this has been offset to some extent by a decline in the average number of hours volunteered. This has occurred at a time when labour force participation has continued to increase.

As might be expected, ABS data show that volunteering rates are higher among women than among men. The difference is confined mainly to the 35-44 year age group – when most female volunteering could be expected to be associated with school canteens etc. People with young children are the group most likely to volunteer regularly, but they spend fewer hours per week volunteering than do people with older children and older people without children.

Again, as expected, the rate of volunteering is higher outside capital cities than within capital cities. But the difference is not huge. The rate for regular participation in voluntary work was 19% in capital cities and 23% outside capital cities in 2006.

The naïve model would not predict that employed people would be more likely to volunteer than unemployed people. For women, although those in full-time employment had the lowest rates of regular volunteering, those who were employed part-time had higher rates of regular volunteering than those classified as unemployed. For men, rates of volunteering for those in full-time and part-time employment were the same and higher than for those who were unemployed.

The most surprising departure from the naïve model relates to donations of money as a substitute for donation of time. I know such substitution does occur, but it doesn’t show up at an aggregate level in the ABS survey data. Volunteers are much more likely to have donated money or contributed financial assistance to someone outside the family in the last 12 months than non-volunteers.

In order to explain non-volunteering we seem to need a model of behaviour that recognizes that volunteers and non-volunteers have different personal characteristics. It seems that non-volunteers tend to have relatively weak links to the community in general. The evidence suggests that they are much less likely to have attended a community event recently. They are also less likely to agree with the proposition that most people can be trusted.

Other posts on volunteering:
This is the third in a series of post on volunteering. In the first post, Shona discussed her experience in a volunteer role in a community playgroup. In the second post I discussed some research on the determinants of volunteering.


Shona said...

One selling point I make for volunteering is that it is a great way to better know the community and make friends. I did it partly for this reason, as well as a host of others. Therefore I found your last paragraph a bit odd.

Winton Bates said...

This is suggesting to me that national surveys don't shed much light on the reasons people have for not volunteering in the eastern suburbs of Sydney.