Incentives are powerful

If you have ever studied economics, you will know that one of its calling cries is that people respond to incentives. Indeed you can even consider economics to be the study of how and why people respond to incentives.

Since public policy is concerned with guiding choices by putting in place incentives, you’d think policy makers would be more in tune with their implications. But policy makers often fail to consider how implementation of public policy can lead to perverse incentives.

An important function of implementation of public policy is to communicate the policy. Often this means making announcements long before the details of the policy initiative is determined. And sometimes just the act of making those announcements can lead to people responding to the policy.

I recently wrote about the unintended consequences of Canada’s expansion of parental leave here. But you might be interested in another feature of this expansion that gets at my point in the blog and is based on my research with Janice Compton on birth seasonality in Canada.

On 12 October 1999 the federal government signaled that it was going to expand the parental leave provisions, but made no commitment as to the date of that expansion. Instead, the expansion was official announced in the 2000 Budget on 20 February 2000 as being effective 31 December 2000. Why are those dates important? The table below shows the quarterly trend in observed births in Canada around this window.

Quarterly Observed Births, Canada, 1997-2004

Year January-March April-


July-Septemberb,c October-December c Total
1997       84,968         92,395         89,500       81,735       348,598
1998       83,424         90,464         88,881       79,649       342,418
1999       81,890         87,875         87,772       79,712       337,249
2000       82,627         86,801         83,173       75,281       327,882
2001       81,350         87,303         86,123       78,968       333,744
2002       79,345         83,719         86,618       79,120       328,802
2003       79,299         85,486         88,856       81,561       335,202
2004       81,583         85,762         87,992       81,735       337,072

Source: Modified from Compton and Tedds, 2016

The date of this Speech was October 1999 and if we use the median gestation period of 40 weeks this accords with July 2000, the beginning of the noted temporary decline in births in Canada. The actual commitment to this expansion was not made until the following Budget, which was announced in February 2000. Again, if we use the median gestation period of 40 weeks this accords with December 2000, the end of the noted temporary decline in births. December 2000 is also notable because the new expanded parental leave policy took effect on 31 December 2000.

While this is certainly not causal evidence, the correlation is notable and suggests that Canadian women responded significantly by curtailing conceptions in the wake of the policy signal. It is also interesting to note that we do not see an immediate spike in births following the policies implementation date, showing what many women already know: it is much easier to cease conceptions than it is to recommence them.

While such a short temporary drop in births may not be concerning, it is an interesting case of how policy announcements themselves can be incentivizing and why policy makers need to pay attention to how their vague policy signals may result in unintended behaviour. Indeed, as the data becomes available, it will be interesting to see if Canadian women had a similar response to the 2015 election announcements related to expanding parental leave from the current 35 weeks to 61 weeks.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s