Of International Trends, Hypotheses and Instrumental Variables

Social trends can be quite dramatic. In the last fortnight, two highly desirable trends have been reported in the UK – a drop in injuries and death due to violence [1] and a drop in the country’s notoriously high teenage pregnancy rates.[2] [3] Observing a trend is one thing, discerning cause is another. The drop in violence has been attributed to decreasing alcohol consumption by young people in the UK. This drop in violence is an international phenomenon, mirrored across the developed world and even, to some extent, in middle-income countries.[4] However, the same change in prevalence of a social problem can have different causes in different places. The drop in violent crime in the USA has been going on for a long time and cannot be attributed to a drop in alcohol use. There are many theories to account for the sustained drop in violent crime in the USA and elsewhere. These include better police tactics,[5] improved security [6] and replacing cash with electronic transactions.[7] Two unusual theories relate to changes in the abortion law and environmental chemical exposure. John Donohue and Steven Levitt attributed the change to liberal abortion laws, which resulted in lower birth rates among under-privileged single parents, and hence a smaller pool of high-risk people 18 years later.[8] One way to test causal inferences from such an association is to seek an ‘Instrumental variable’ – a change that is thought to be independent of the outcome variable of interest. The effect of introducing or removing capital punishment on homicide rates is an example. In the case of violent crime, Donohue and Levitt compared the year in which individual States liberalised abortion laws – which they regarded as an instrumental variable – with the year in which violence started to decline in that particular state. The drop in crime followed the liberalisation in law after a set interval. This trend has been observed in other countries – Canada,[9] Australia [10] and Romania.[11] In the meantime another provocative hypothesis linked the drop in lead levels in petrol to the decline in violence and there is evidence that lead can predispose to violence,[12] [13] with some even attributing the downfall of Rome to the ubiquitous use of lead pipes in the Empire, though a recent paper concluded the likely concentration of lead would be unlikely to represent a major health risk.[14] A study linking the point in time where crime rates started to decline to control of lead levels in the environment would be revealing – as usual, please respond if you know of, or are planning, such a study. The drop in violent crime came later in Europe, or not at all,[15] and would appear to have different antecedents, of which binge-drinking is one. The UK drop in teenage pregnancy rates might be the result of government action. Certainly a concerted and rather expensive national program was put in place, starting in about 2000. Further, the experimental evidence, taken in the round, supports the effectiveness of educational interventions to increase contraceptive use and decrease pregnancy rates without reducing the frequency of sexual encounters.[16] [17] There has been a drop in teenage pregnancy rates across OECD countries [18] and maybe the amount of education on contraception and relationship management has been upgraded in schools across the continent? — Richard Lilford, CLAHRC WM Director. References:

  1. Sivarajasingam V, Page N, Morgan P, Matthews L, Moore S, Shepherd J. Trends in community violence in England and Wales 2005-2009. Injury. 2014; 45(3): 592-598.
  2. Arie S. Has Britain solved its teenage pregnancy problem? BMJ. 2014; 348: g2561.
  3. Office for National Statistics. Conceptions in England and Wales, 2012. 2012.
  4. van Dijk J, Tseloni A, Farrell G. The International Crime Drop: New Directions in Research. Hampshire: Palgrave Macmillan; 2012.
  5. CBS News. Chicago police credit new tactics for drop in homicides. Feb 7 2014.
  6. MacDonald JM, Klick J, Grunwald B. The Effect of Privately Provided Police Services on Crime. Faculty Scholarship. Paper 430. 2012.
  7. Wright R, Tekin E, Topalli V, McClellan C, Dickinson T, Rosenfeld R. Less Cash, Less Crime: Evidence from the Electronic Benefit Transfer Program. NBER Working Paper 19996. 2014.
  8. Donohue JJ, Levitt SD. The impact of legalized abortion on crime. Quart J Econom. 2001; 2: 379-420.
  9. Sen A. Does Increased Abortion Lead to Lower Crime? Evaluating the Relationship between Crime, Abortion, and Fertility. BE J Econ Anal Poli. 2007; 7(1).
  10. Leigh A & Wolfers J. Abortion and Crime. AQ: J Contemp Anal. 2000; 72(4): 28-30.
  11. Pop-Eleches C. The Impact of an Abortion Ban on Socioeconomic Outcomes of Children: Evidence from Romania. J Pol Econ. 2006; 114(4): 744-73.
  12. Nevin R. How Lead Exposure Relates to Temporal Changes in IQ, Violent Crime, and Unwed Pregnancy. Environ Res. 2000; 83(1): 1-22.
  13. Nevin R. Understanding international crime trends: The legacy of preschool lead exposure. Environ Res. 2007; 104: 315-336.
  14. Delile H, Blichert-Toft J, Gorian J-P, Keay S, Albaréde F. Lead in ancient Rome’s city waters. PNAS. 2014. [Online].
  15. Buonnano P, Drago F, Galbiati R, Zanella G. Crime in Europe and in the US: Dissecting the ‘Reversal of Misfortunes.’ Econ Policy. 2011; 26(67): 347-85.
  16. Kirby DB, Laris BA, Rolleri LA. Sex and HIV Education Programs: Their Impact on Sexual Behaviors of Young People Throughout the World. J Adolesc Health. 2007; 40(3): 206-17.
  17. Harden A, Brunton G, Fletcher A, Oakley A. Teenage pregnancy and social disadvantage: systematic review integrating controlled trials and qualitative studies. BMJ. 2009; 339: b4254.
  18. Organisation for Economic Development. Share of births outside wedlock and teenage births, 2013. 2013.

2 thoughts on “Of International Trends, Hypotheses and Instrumental Variables”

  1. Concerning the paper cited in 14, authors clearly said that “The inferred increases of Pb in the water of the
    Roman distribution system unquestionably attest to general lead
    pollution of Roman drinking water but the Pb concentrations at
    issue are unlikely to have represented a major health risk.”.

    Therefore, your citation of this study is inappropriate since it says exactly the opposite. To improve your paper, you can either delete this wrong citation or change your sentence.

    A more appropriate paper about this health hazard at Rome during Roman times has been published by the same first author in the journal Medecine/Science.

    A careful reader

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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