GDP isn't just adding up all the nice things

People keep wanting GDP to be something it isn't.

The only thing that GDP is is a measure of the final value of goods and services that trade in markets. That's it.

There are all kinds of good things that are not in GDP.

High among those good things is the value of household production that does not trade in markets.

This is standard fodder in principles and intermediate-level coursework. If you have a two-parent household, with one working outside the home for wages and the other working inside the home, then the value of in-home production does not count toward GDP. If the one parent starts paying the other one, then GDP goes up - even though absolutely nothing has changed.

If I make a sandwich at home, the value that I add to the ingredients by my labour is not counted toward GDP. If I sell the sandwich to myself, like if I were owner-operator of sandwich shop, it would be - again, despite there still being no change in the real economy.

If I babysit the neighbour's kids at our place for free in exchange for their babysitting ours, it doesn't contribute to GDP. If I paid them, and they paid me, it would be.

This is all standard stuff.

And there are some real problems caused by this, but they have nothing to do with GDP. They have to do with tax. But we'll come back to that.

ANU's Julie Smith was on RNZ today arguing the case for including the value of breastmilk in GDP. She argues that the value of unpaid household services should be in GDP. She's right that there is a problem if we're setting GDP growth rates as a target and we're ignoring that increases in female labour force participation has been at a cost to unmeasured but valued household production. Patricia Apps made similar points in her keynote at the NZAE meetings this year.

But there's good reason for keeping GDP as it is, and just being careful about how it's used.

Let's start thinking about all of the unpriced non-market activities that go on and that could, alternatively, be provided within markets.

Parents provide a lot of services for their kids, from chauffeuring to tutoring, and from homecare to mentoring. People can hire Ubers, and tutors, and home-care workers, and life coaches. Valuing all of those services would be tricky. And it would be pointless if GDP numbers were being used in ways that they should be used.

Except, that is, when it comes to tax.

I'd raised this as question during Patricia Apps' keynote at the NZAEs. People thought I was joking as reductio, but it's a serious point - and I expect a very real distortion. Just one that's probably not worth worrying about because trying to fix it would be even worse.

The distortion is as follows.

If you have a two-income household, both earners pay income tax on their earnings. And they pay GST for the services that they have to buy-in to help around the house, if they're buying in services to help with the lost time for home production. And the workers providing those in-home services pay tax on that income.

In-home services are paid for out of after-tax income, are subject to GST, and the worker takes home an after-tax income.

That builds a substantial tax wedge encouraging the in-sourcing of a lot of services, and distorting activity and formal labour force participation. There is a substantial tax advantage to having one partner stay home and provide untaxed services rather than be out in the formal paid workforce.

I recall stories about, when top marginal tax rates here were a lot higher, econ faculty doing a lot more of their own home renovation work. The tax wedge mattered. It's the same kind of problem.

Effectively, single-earner families are tax dodgers. No GST is paid on the in-home services provided by the stay-at-home parent. No income tax is paid on the monetary transfers to the stay-at-home parent from the in-work parent.

So I'd asked Apps whether, if we wanted to be really serious about addressing the value of household production, we shouldn't be taxing single-earner families based on the value of the household services implicitly provided. I don't think she'd thought about the problem that way before.

Of course, down this path lies madness. There are plenty of services provided between couples that do also trade, one way or another, in markets - legally in New Zealand, illegally in other places. But we'd all recognise it as insane to wish to impose GST on the imputed value of those activities - or to start having Stats NZ ask couples how often they had sex, put a dollar value on it, and start adding it into the GDP statistics. It sounds nuts and all, but as sex work is now legal, every visit to a brothel counts toward GDP (and attracts GST and income tax), while tax-dodging black market activities in the bedrooms of the nation do not.

Better I think to just keep GDP as it is, and recognise its limitations for policy purposes.

Risky diagnoses

When you're cautious in taking sexual risks, you help both yourself and your partners. The former effect can be purely selfish optimisation. The latter could be due to other-regarding preferences in relationships where you care about the other person, or just a positive externality.

Why does this matter? Consider what happens if more HIV testing is funded. If you're behind the veil and don't know about your status or your partner's, prudence dictates some caution. If you're tested, you know your status but don't necessarily know your partner's. If you're tested and wind up being negative, then the returns to prudence are higher, as you know with certainty that you aren't already infected so you can make things worse, but you might also think that the risks you've taken so far are safer than you'd thought. And how much weight people put on the risk they impose on partners is hard to tell: if you find out you're positive, you either reduce caution if partners' utility doesn't weigh heavily, or increase caution if others' utility counts.

And so what people do on getting a test result is an empirical question.

Enter Erick Gong. He finds that ..., well, scratch that. I'll just quote from his introduction as it's rare to see this kind of clarity in academic writing. Bottom line: if you're going to fund free testing, couple it with funding for anti-retro virals so that when people find out they're positive, they do less harm to others.

I use data from the Voluntary Counselling and Testing (VCT) Efficacy study conducted in Kenya and Tanzania, which randomly assigned people into HIV testing and followed up with them six months later (The Voluntary HIV-1 Counselling and Testing Efficacy Study Group,2000). I construct a measure of people's beliefs about their HIV status before getting tested using questions on the baseline survey. To measure risky sexual behaviour, I use biological markers that are not susceptible to self-reporting bias. Data are collected on newly contracted infections of gonorrhoea and chlamydia (henceforward known as ‘sexually transmitted infection’ or ‘STI’) that occur during the study.5 An STI only results from unprotected sex with someone who has an STI and serves as an objective measure of risky sexual behaviour. The random assignment of testing enables me to identify the effect that HIV tests have on sexual behaviour conditioned on prior beliefs of HIV infection.

My findings suggest that HIV tests have the largest effects on risky sexual behaviour when test results provide unexpected information to an individual. I find that people surprised by an HIV-positive test (i.e. those who believed they were at low risk for HIV before testing and learn they are HIV-positive) have a 10.5 percentage point increase in their likelihood of contracting an STI compared to an HIV-positive control group who had similar beliefs of HIV risk but were untested at baseline.6 I interpret this increase in contracting an STI as an indication that those surprised by an HIV-positive test increased their risky sexual behaviour – an unintended consequence of testing. I estimate that these types on average increased their number of new partners by about 2.4 over a six-month time frame. People surprised by an HIV-negative test (i.e. those who believed they were at high risk for HIV before testing and learn they are HIV-negative) have a 5 percentage point decrease in the likelihood of contracting an STI compared to an HIV-negative control group with similar beliefs of HIV risk but were untested at baseline.7 This decrease in the likelihood of contracting an STI suggests that those surprised by HIV-negative tests decrease their risky sexual behaviour. Both of these results indicate that when people make decisions about risky sexual behaviour, self-interests dominate altruistic preferences. People who discover they are HIV-positive no longer have any incentive to practice safe sex (i.e. ‘nothing to lose’), while those who learn they are HIV-negative face greater incentives to avoid risky behaviour. Finally, when HIV test results agree with a person's beliefs of HIV status, the effects of testing on STI likelihood are not statistically different from zero. This is consistent with an economic model where the behavioural responses to HIV tests are greatest when they provide unexpected information.

I use the empirical results described above and combine them with a simple epidemiological model to simulate the short-run effect of rolling out HIV testing in an urban setting. While this exercise inherently requires a set of strong assumptions, and hence the results should be interpreted with caution, it does address an important policy question. I use the distribution of beliefs of HIV risk and actual HIV status from the Demographic Health Surveys in Kenya, Mozambique and Zambia – all three countries faced with a generalised HIV epidemic. I find that in the cases of Kenya and Zambia, testing leads to declines in new infections, while testing leads to an increase in infections in Mozambique. However, when ARVs are provided at an earlier stage in the infection, testing leads to large reductions in HIV infections in all three countries. Since ARVs greatly reduce the infectivity of HIV-positive individuals, the aggressive provision of ARVs can mitigate the risks posed by HIV-positive individuals who increase their risky sexual behaviour after testing.8

The other particularly interesting bit: surprise negatives yield reductions in risk-taking.

I'd missed this when it came out in 2015. I thank Ole Rogeberg for the pointer.

Inequality and unhappiness

If income inequality is bad, in part, because those on lower incomes feel bad about their absolute levels of income when others are doing better, what is appropriate policy on this one?

When people grade their sex lives relative to their peers, it takes a toll on their overall happiness, not just satisfaction with their sex lives. Tim Wadsworth, an associate professor of sociology at the University of Colorado Boulder, found that falling behind other people’s sex lives can lower overall happiness. For every level of sexual activity (two to three times a week, once a week, two to three times a month, etc.) that people were behind the actual average for their peers, they were 14 percent more likely to describe themselves as “not too happy” rather than either “pretty happy” or “very happy.”²

And just as unhappiness with income inequality is more driven by perceptions of inequality than by the data, so too for inequality in sex lives:

Inaccurate perceptions about what counts as normal sexuality are widespread. In sociologist Michael Kimmel’s book “Guyland: The Perilous World in Which Boys Become Men,” he found that male college students assumed about 80 percent of their classmates had sex on any given weekend. The real number was closer to 5 percent to 10 percent. Kathleen Bogle, the author of “Hooking Up: Sex, Dating, and Relationships on Campus,” also found in her interviews that students consistently overestimated the amount of sex that others were having.The result is a reverse Lake Wobegon effect: Everyone is below “normal.”

As always, which inequalities matter for policy is at least as interesting as the stats on any particular margin.

IUDs and teen pregnancy

Better access to IUDs at subsidised family planning clinics reduces teen birth rates, says a new NBER working paper by Jason Lindo and Analisa Packham.

From their abstract:

Despite a near-continuous decline over the past 20 years, the teen birth rate in the United States continues to be higher than that of other developed countries. Given that over three-quarters of teen births are unintended at conception and that over a third of unplanned births are to women using contraception, many have advocated for promoting the use of long-acting reversible contraceptives (LARCs), which are more effective at preventing pregnancy than more commonly used contraceptives. In order to speak to the degree to which increasing access to LARCs can reduce teen birth rates, this paper analyzes the first large-scale policy intervention to promote and improve access to LARCs in the United States: Colorado's Family Planning Initiative. We estimate its effects using a difference-in-differences approach, comparing the changes in teen birth rates in Colorado counties with Title X clinics (which received funding) to the changes observed in other US counties with Title X clinics. The results of this analysis indicate that the $23 million program reduced the teen birth rate by approximately 5% in the four years following its implementation, providing support for the notion that increasing access to LARCs is a mechanism through which policy can reduce teenage childbearing.

I'd love to know whether the programme also had effects on STD rates. Where the price of unprotected sex goes down, you should get more of it - as Klick and Stratmann found. That's hardly a reason not to provide access, just something to consider as an offsetting cost potentially in the mix.

Relationship-specific investments

I spend a week on sex, love and economics in my Econ & Current Policy Issues class. I there note that one function of marriage is the encouragement of relationship-specific investments.

Suppose that you have two broad classes of action for an hour's worth of free time. You can invest in an activity that will increase your value to other potential partners, or you can spend your time on something that wouldn't be noticed by outsiders but provides a lot of value within the relationship. If you think that your relationship will be short-lived, you do more of the former; if you think it'll last, you do more of the latter in hopes of reciprocation. The principle's pretty general: employers might not want employees to think they're for the chopping-block if they want employees to pay attention to those bits of the job that matter but aren't measured. Any time you want to induce effort in the provision of relationship-specific goods, you need some assurance of a longer-term commitment.

Today's bit of evidence comes from a new survey of dating data. Women in longer term relationships are more satisfied with some aspects of the partner's performance than are those in short-term hook-ups. There will be a bit of sampling bias built-in here: those in long-term relationships are those who've found good matches while those in short-term ones are less likely to have found their match. But the data's consistent with what we'd tend to expect from base theory.

The Social Costs of Abstinence

Suppose that the correlation between one's sex life and earnings were actually causal, and worked from sex to income rather than the other way round. What correlation? This one:

Having an active sex life may make you happier, healthier and wealthier.
A new study reveals that people who had sex four or more times a week earned more money than their counterparts who weren't as lucky.
"People need to love and be loved (sexually and non-sexually) by others. In the absence of these elements, many people become susceptible to loneliness, social anxiety, and depression that could affect their working life," study author Nick Drydakis, an economics lecturer at Angila Ruskin University in Cambridge, England, said to CBSNews.com by email.
The discussion paper was published in July by the Institute for the Study of Labor,an economic research institution, in Germany.
Drydakis said he was interested in the topic because of previous studies linking sexual activity with extroversion traits (including being sociable, outgoing and energetic) and good health. In addition, good health has been linked to higher wages. A 2009 Brazilian study also showed a connection between higher wages and a more active sex life.

We can imagine some causal mechanisms that could run from sex to income. Happier people could be more productive at work. Or the cardiovascular benefits could yield better health and then consequently greater productivity. The authors do use a two-stage estimation procedure to try to isolate causality: they try to instrument for sexual activity, so it's at least more plausibly causal than much of what goes on in the public health literature.

If we follow the standard line in public health of assuming correlations are causal and in the "right" direction, and of ascribing as social all things private, we have to then worry about the social costs of abstinence. Those having too little sex earn less and so must be less productive. Those productivity costs reduce output and reduce tax revenue. And if it's working through a health channel, they also impose costs on the public health system.

The policy consequences are left as an exercise for the reader. But note that if you're recommending subsidies, you might need to offset the STD costs on the public health system via complementary regulations around health testing and public disclosure of who has what. This may seem like a violation of privacy, but can we really make rational decisions without perfect information?

[The should-be-obvious caveat: entire post subsumed within a "reductio" tag.]

Underlying type

Does alcohol or drug use lead to a whole pile of other risk-taking activities, does some common underlying risk preference determine both substance abuse and other risky behaviours?

The Dunedin Longitudinal Survey group finds evidence that women with more sexual partners are more likely to later report substance dependence disorders than those with fewer partners. Women having had more than 2.5 sexual partners per year between the ages of 18-20 are 9.6 times more likely to report substance dependence disorders at age 21, adjusted for prior disorder incidence. Women aged 26-31 having had more than 2.5 partners per year are 17.5 times as likely to report substance dependence disorder at age 32. Similar patterns held among males, although the risk ratios were much smaller. 

The explanation for the relationship is likely to be complex. Four possibilities are proposed. First, sexual risk taking and substance use may be part of the cluster of risk taking behaviors common in adolescence and young adulthood (Arnett, 1992; Boyer et al., 2000; Caspi et al., 1997; Desiderato & Crawford, 1995; Donovan & Jessor, 1985; Taylor, Fulop, & Green, 1999). For instance, people who are impulsive may be more likely to engage in both activities and, consequently, more likely to become substance dependent. Second, occasions of substance use are opportunities for sexual behavior because of its disinhibitory effects and lack of accurate perception of risk (Crowe & George, 1989; Fromme, D’Amico, & Katz, 1999). Weinhardt and Carey (2000) have suggested, in a review of event-level research on this topic, that the association, especially with condom use, is also complex. Thirdly, shared context may be an important factor, insomuch as young people are likely to meet new sexual partners in situations where alcohol is served. These settings might encourage sexual behavior and facilitate multiple partnering.

The fourth intriguing possibility is that it is something about having multiple sex partners itself which puts people at risk of substance disorder. For instance, it may be due to the impersonal nature of such relationships. Or, it might be that multiple failed relationships create anxiety about initiating new relationships. Self "medication" with substances may be one way of dealing with this interpersonal anxiety (Khantzian, 1997; Stoner, George, Peters, & Norris, 2006).

They also note the studies showing that alcohol use correlates with more risky sexual practices; I didn't see reference to the one suggesting drinking was associated with more positive consequences of sex.

I'd also worry about cohort attrition effects: if men and women who are more psychologically stable are more likely to get married before the age of 30, then the pool of women reporting >2.5 sexual partners per year* between the ages of 26-31 is probably different from the pool of women reporting the same numbers in their early 20s [recall that the Dunedin study follows a cohort born in 1972-1973]. 9.5% of women reported 2.5+ partners per year at 18-20; that dropped to 4.5% by age 21-25 and to 1.7% - 8 women - by age 26-31.

I wish that the Dunedin study had some calibrated measure of risk tolerance, like the Holt and Loury measure, as well as a measure of individual discount rates. I would love to see pinned down what portion of risk-taking behaviour comes down to heterogeneity in individual risk tolerance, what portion comes down to that things with longer term costs might be disproportionately preferred by those who avoid the tyranny of the later-self, and what portion might be due to amplification effects where doing one risky thing actually does make you more likely to do another risky thing.

But the takeaway here is that studies suggesting drinking is associated with riskier sexual activity might well worry about reverse causation or common underlying causes.

* No, you can't have half a sexual partner. They're asked number of partners and that's averaged across the age range for that respondent. 

Causes and consequences

One thing that struck Matt and me when going through the alcohol literature was co-morbidity between heavier alcohol use and a whole pile of other bad things. We wrote:

We also worry that the aetiological fractions used may not account adequately for comorbidity between alcohol use and pre-existing mental disorders. The aetiological fractions used ascribe between twenty-five and thirty percent of male suicides to alcohol; in other words, if alcohol disappeared, the suicide rate would drop by more than a quarter for adult males over the age of twenty. As alcohol use can often be a form of self-medication among those with mental illness, whether alcohol plays that substantial a
causal role in suicides is debatable. Ross (1995) finds that more than half of those with an alcohol disorder have a lifetime comorbid psychiatric disorder. Among subcategories for which data is presented, alcohol abusers have rates of mood disorders and anxiety disorders 2.3 and 1.7 times that of non-abusers. While 9% of alcohol abusers report antisocial personality disorders, only 0.6%  of non abusers report such disorders. The Mental Illness Fellowship of Australia (2005) notes that those with bipolar disorder are eleven times more likely to engage in harmful drug or alcohol use than is the general population. Kessler et al (1997) find that those with long term alcohol abuse or dependence not only have a high probability of also exhibiting another mental disorder but also that comorbid DSM-IIIR disorders tend to predate alcohol use disorders.  

Some people with pre-existing disorders self-medicate with alcohol and consequently are better off than they would have been but worse off than average; their variance from average outcomes is counted as a cost of alcohol. Some people with pre-existing disorders self-medicate with alcohol and consequently are worse off than they otherwise would be; the total amount of their variance from average outcomes is counted as a cost of alcohol rather than only the incremental worsening from their individual baselines. This stuff isn't easy. But the direction of the bias is pretty clear.

Emma Hart at Up Front [not guaranteed safe for work] points to more work in similar vein: bisexuals are more likely to have problems with binge drinking. And it's because of underlying social pressures. She writes:

Here’s a clue, guys: it’s not the drinking, it’s the why they’re drinking. Take a lesson from gaynz.com, and maybe work out why I link to so many stories there instead of at your place: Exclusion Leading Some Bi Youth to Binge Drink. The drinking is not the problem, it’s one of the symptoms of the problem. 

She then quotes from the study's surveys of young binge-drinking bisexuals:

I drink more when I’m under high stress, when I’m stressed out, and maybe sometimes at parties when, after conversations with people, where they want to know, no one gets the bi thing. It’s really hard to explain. Quite a bit because you get people who want to know why you are not lesbian, why you are not straight, and I kind of feel that, it’s slightly easier to be one or the other, like I envy some of my friends who are gay, I’m like you know who, you know you’re there and no one questions it. But I get questioned all the time, and I find that frustrating sometimes.
...
Arahia: You kind of drink more so you can say the next day: “Oh, I was just drunk, you know. It didn’t mean anything really.” Sometimes it does, sometimes. But if you wake up the next morning with a huge hangover, you can say to the person: “Oh god, it didn’t mean anything. I was just so wasted.”

Fiona: “Didn’t mean to grope you. I was just drunk.”

Arahia: It is such a good excuse.

Fiona: And I think bi people definitely use it as more of an excuse than any other sexual orientation.

Sometimes, use of alcohol is inframarginal to whatever other behaviour we're measuring, whether crime or sex. Sometimes, alcohol use helps people to get over their inhibitions and do utility-enhancing things. And, sometimes alcohol use leads people to make mistakes. We overstate the costs of alcohol when we assume that alcohol's role is always causal and always negative, or when we assume away the positive consequences.

Ban the Cup

The latest research out of the University of Otago warns of the dangers of major sporting events.

It seems Rugby-World-Cup-related sex is dangerous. I can't access the original article as the journal is currently returning a 503 error code, so I'll have to rely on the newspaper reporting.

According to the New Zealand Herald, folks attending Sexual Health Clinics around the time of the Cup were surveyed. Those reporting having had RWC-related sex, about 7% of the sample, had higher risk of STDs, often reported having consumed alcohol before the act [the DomPost says 70% had consumed alcohol], and rarely reported having used condoms. From this, the authors argue for reduced promotion and availability of alcohol around future large sporting events.

Again, I haven't access to the article. But a few things come to mind.

First, sample selection is an awfully large problem here. 151 people attended a Sexual Health Clinic after having had RWC-related sex. How many people had sex after a fun night out watching the matches, didn't wind up with suspicious itches or discharges, and so didn't go to visit a Sexual Health Clinic? What was the rate of alcohol or condom use among those who failed to show up at a Sexual Health Clinic? How many of them simply had a great time without negative consequences? Recall that alcohol consumption correlates with positive sexual experiences.

A second sample-selection issue is that folks reporting RWC-related sex are probably really reporting "sex after hooking up at a RWC party or event", which won't be that different from "sex after hooking up at any party or event". And, in that case, again, is the fault with the alcohol, or is it that people going out and wanting to have a good time are more likely to drink and also more likely to think that hooking up is fun? Recall that the answer to "Do you like beer?" is a significant predictor of whether someone will have sex on the first date. The omitted underlying variable is then likely some combination of sensation-seeking, risk preference, and hedonism. Unless you control for that underlying heterogeneity, you're going to draw some awfully misleading conclusions from straight correlations among the various outcomes of that underlying agent-type.

Finally, here's how the Herald describes "RWC-related sex":

People who had RWC-related sex were defined as New Zealanders who had sex related to the RWC or other associated events, New Zealanders whose sexual event leading to the clinic visit was with an overseas visitor primarily in New Zealand for the RWC, and individuals visiting New Zealand primarily for the RWC.

So, basically then, people who report having hooked up with a tourist, tourists hooking up with locals, and folks hooking up at parties are more likely to engage in riskier sexual activities, many of them have had a few drinks, and some of them show up at Sexual Health Clinics. Policy conclusion: crack down on alcohol.

About 133,000 tourists showed up for the RWC; rather a few locals attended RWC events. 151 people showed up at Sexual Health Clinics afterwards. Suppose that ten percent of the tourists here for the RWC had some kind of RWC-related sex with locals, each only having one local partner. That's then about 26,500 people having had RWC-related sex; add to that locals hooking up with each other. I have no clue how close to right those estimates are. But I'd be pretty surprised if that 151 were more than 1% of the total RWC-related hook-ups. And this is the basis for an alcohol crackdown?

Otago remains an interesting place.

9/11 as an Instrument

It's hard to tell in cross-sectional data whether depression is cause, consequence, or spurious correlate of risky sexual activity. Susan Averett and Yang Wang try to figure it out in the latest AER. Their trick? AddHealth data collected over a period that spanned 9/11 that lets them use the terrorist attacks as an instrument for depression.

Some summary statistics:

Our full sample consists of 4,151 women aged 18 to 26 years. Most of our respondents are white (68 percent) or African American (25 percent). The average years of education are 13.5 and the average age is 21 years. Thirty-five percent smoked for at least 30 days in the past year, and during the year before interview 72 percent of these women drank alcohol and 31 percent used marijuana. Selfrated health averages 2.05, which is slightly worse than “very good.” The mean CESD score for the nine questions asked in the Wave III 95 percent of these women are sexually active in their relationships. Sixty-seven percent and 11 percent of them engage in oral and anal sex, respectively. Fewer than 2 percent of the women in our sample report using condoms.

An active sample. And, an active sample that showed a sharp jump in depression scores in the period right after 9/11. They then use 2SLS with 9/11 as instrument and find that OLS estimation underestimates the effect of depression on participation in risky sexual activity.

So a one point increase in the CESD (depression) score correlates with a 0.4% increased likelihood of vaginal sex by OLS, but a 2.1% increase by 2SLS - and a 1.9% reduction in the likelihood of using a condom. 

It's worth keeping this kind of result in mind when we hear findings that alcohol use correlates with risky sexual behaviour [and, let's not forget, with more positive consequences of sexual experiences]. If depression correlates with heavier drinking, then it's pretty easy to conflate the effects of alcohol with a covariate, like depression, that drives both drinking and risky sexual practices.