Fresh water is still not a public good

Earlier in the month, this New Zealand Herald article caught my attention:

Will 2021 be the year the world really values water?

If Wall Street sets the tone, it will be.

For almost 230 years, agricultural commodities have been bought and sold in New York's finance district.

And now the Nasdaq stock exchange, which celebrates 50 years of activity next month, has put a price on our most vital substance.

Water contracts for five water districts in drought-prone California are being bought and sold.

The new water futures contract allows buyers and sellers to barter a fixed price for the delivery of a fixed quantity of water at a future date.

In December, for the first time, water futures for drought-hit California districts are also being traded on the floor of the world's second-biggest market.

The concept, which has been mooted for decades, finally came about in December.

That all seems pretty sensible so far. A futures contract allows the water user to 'lock in' a future price for water, reducing the risk that they will be caught out by unanticipated future price rises. However, then we get to this bit:

The move was quickly criticised by public health specialists.

Pedro Arrojo-Agudo, the United Nations' special rapporteur on the human rights to safe drinking water and sanitation, was direct in his opposition.

"You can't put a value on water as you do with other traded commodities.

"Water belongs to everyone and is a public good. It is closely tied to all of our lives and livelihoods, and is an essential component to public health.

"Water is already under extreme threat from a growing population, increasing demands and grave pollution from agriculture and mining industry in the context of worsening impact of climate change."

There are several problems with this line of reasoning. First, water is not a public good. As I noted in this 2017 post on the same topic, by definition a public good is a good that is non-rival (where one person using the good doesn’t reduce the amount of the good that is available for everyone else) and non-excludable (where the good is available to everyone if it is available to anyone). Unless you live in Ankh-Morpork [*], the first condition clearly doesn't hold - one person using fresh water leaves less available for everyone else. Fresh water is rival, not non-rival. A good that is rival and non-excludable is a common resource. However, that isn't the case for California water, which is allocated through water rights. You must have water rights to draw water, making water excludable. A good that is rival and excludable is a private good. Fresh water in California is, by definition, a private good.

Second, although the availability of fresh water may be under threat from increasing demand, putting a price on water is a solution to that problem, not something that exacerbates the problem. With a price on water, the price will dictate how much water people use. In times of drought, the price of water should rise (unless the price is controlled by the government, like it is in Auckland), and people will use less water. Goods that are scarcer have higher prices - as water gets scarcer, the price will rise. This is a means of better managing scarce fresh water supplies, not some nefarious plot to take water out of the hands of the people.

Third, water already has a price in California. To draw water, you need water rights, and those water rights cost money. The only thing changing is that a futures contract has been introduced, so that water users can better manage the future uncertainty of water prices. If a drought is expected in the future, then the price of the futures contract will rise. Water users will have an incentive to act now to ensure that their future water use will be lower. That probably makes water use more efficient.

You don't have to be a market fundamentalist to realise that prices can actually help. In the case of fresh water, the alternative is a free-for-all, where the water supplies will almost certainly be depleted faster.

*****

[*] Terry Pratchett noted that the water in the Discworld's largest city must be very pure, because of the number of kidneys it had already passed through.

Read more:

• Fresh water is not a public good
• Why we need tradeable fresh water usage rights

Climate change and population control

I had my first ever article published in The Conversation this morning, on population control and climate change. I encourage you to read it.

The process of writing and getting the article published has been an interesting experience. It started as a request from the University's communications team to write something 'for the University website'. Once I had written a draft though, it became clear that I was being subjected to what I am sure was a pretty unsophisticated bait-and-switch, when the comms people recommended I pitch the article to The Conversation. The article appears to be doing what they intended though, and my email has been pinging constantly all day as the article attracts comments (129 comments so far, and counting - for context, that is more comments than all other articles by University of Waikato staff have attracted in the last month).

Anyway, there isn't much space to write in depth about the issues in the 800 words that The Conversation allows, and the original 1200 words I drafted got pared back substantially. The main issue I wanted to highlight is that, when some people argue that we should curb population growth to reduce carbon emissions (or climate damage more generally), there are a number of key issues that need to be considered. First, there is an ethical or moral issue, as Ross Douthat discussed in this article back in November. As I noted in my article:

If our concern about climate change arises because we want to ensure a liveable future world for our grandchildren, is it ethical to ensure that pathway is achieved by preventing some grandchildren from ever seeing that world because they are never born?

Second, it's not clear that government-enforced population control is even necessary, because:

All high-income countries currently already have below-replacement fertility, with fewer children being born than are necessary to maintain a constant population.

What does that mean? I took my starting point as the Kaya Identity, which breaks carbon emissions exactly into four components:

1. Population
2. GDP (or production) per capita;
3. Energy use per unit of GDP (or production); and
4. Carbon emissions per unit of energy use

To reduce emissions, we would need to reduce one or more of those components. Taking into account that the first two components are growing, if we can't easily reduce population and we are unwilling to reduce economic growth, then we need to reduce the energy intensity of the economy or the carbon intensity of energy production. And if we can't do either of those things (and, to be honest, we haven't done a good enough job so far), then we would have to undertake some challenging conversations about the first two.

Anyway, against my better judgement I have been following along the discussion in the comments, and a couple of very valid points have been raised. So, I thought I would make some comments about them here.

First, the moral or ethical issue that I raised has been challenged. Some people have missed the point that it is a philosophical question and so, while the answer might seem obvious on the surface, the reasoning is less so. However, aside from the philosophical question, the moral issue I raised does take a human-centred view of the world, whereas if you take a planet-centred view there is no such issue. That is, would we save the planet for our grandchildren at all, or would we save it for the planet itself? That is an excellent point.

Second, I raised China as being the only country to have undertaken a successful policy of population control. Some commenters raised family planning programmes, but I would argue that family planning doesn't have at its heart a goal of population control - it is about empowering parents (particularly women) in their choices about family size. It's still a good point though, and there is a huge unmet need for family planning (particularly in developing countries and in underserved communities in developed countries) that, if resources were applied, could reduce future population. Increasing female education also has the effect of lowering fertility rates but again, it's not a policy a government implements explicitly for its population control effects.

The minimum wage and teenage birth rates

The disemployment effects of minimum wages continue to be debated. If a minimum wage does reduce employment though, we would expect it to have the greatest effect on young people and workers with low human capital. So, when a minimum wage is increased, we might expect some teenagers to be disemployed, and those that are not will receive a higher income.

In a new article published in the journal Economics Letters (sorry, I don't see an ungated version), Otto Lenhart (University of Strathclyde) looks at the effects of the minimum wage on teenage birth rates. As Lenhart notes, there are a number of reasons to believe that the minimum wage might have an effect on teenage pregnancy, but not all of the potential effects work in the same direction. Lower time spent working frees up time for 'leisure activities', but if it also lowers income it reduces the resources available for raising children. On the other hand, higher income for those that remain employed provides better access to health resources, including health messages and contraception. It also raises the opportunity cost of time that could be employed in raising children, or in 'leisure activities'. So, theory doesn't provide a definitive answer as to whether a higher minimum wage would decrease, or increase, teenage pregnancy.

Lenhart uses U.S. state-level data on teenage (15-19 years) birth rates and minimum wages over the period from 1995-2017. The minimum wage is measured as the ratio to the state-level median (hourly) wage. Using a difference-in-differences approach, he finds that: 

...increases in minimum wages are associated with reductions in teen birth rates. Using one-year lagged minimum wages, I find that a $1 increase in the effective minimum wage reduces state teen birth rates by 3.43 percent (p<0.05). While slightly smaller in magnitude, the negative and statistically significant effect remains when including state-specific time trends.

Interestingly, the effect is similar but smaller for women aged 20-24 years, and smaller again for each five-year age group above that (becoming statistically insignificant for women aged 40-44). Also interesting is that the effect is only statistically significant for states that have an earned income tax credit (EITC).

Lenhart doesn't really explore this latter finding too much, but I think he should have. It's not clear to me that is helps us understand the mechanisms through which this change occurs. EITCs increase employment and income, while minimum wages increase income (for some), but reduce employment (at least, that's still my overall conclusion on the evidence in the long-running minimum wage debate). If a higher minimum wage leads to lower teenage pregnancy, but only in states with an EITC, then it is tempting to say that this is all a result of higher income. However, in the U.S., I believe that the EITC is only paid to working parents. So, the higher income as a result of the EITC only occurs in the case of having a child, so if teenagers are having less children, then they won't benefit from the EITC. A higher minimum wage reducing teenage birth rates only in states that have an EITC doesn't make a whole lot of sense to me. Unless I'm really not understanding this.

Clearly, there is more work to do on this topic. It isn't as clear as the effect of minimum wages and EITCs on criminal recidivism.

Is Aaron A. Aardvark an exporter?

Following on from my post earlier this week about solving alphabetic discrimination in economics, it appears that economics isn't the only place where such discrimination exists. A recent article by Hua Cheng (Peking University), Cui Hu (Central University of Finance and Economics, China), and Ben Li (Boston College), published in the Journal of International Economics (no ungated version, but there are some presentation slides available here), looks at the case of exporters in China. This is a really nice paper because of the way they tried to identify the alphabetic effects, by exploiting the fact that Chinese companies have two names and hence two different alphabetic rankings - one in English and one in Chinese. As Cheng et al. explain:

The first character has a character rank determined by the Chinese lexicography (number of strokes), and a romanization rank determined by the English lexicography (modern English alphabet). By interacting (i) the character rank with every foreign market's language proximity to Chinese, and (ii) the romanization rank with every foreign market's language proximity to English, we formulate a dual difference-in-differences strategy to estimate how a lexicographically earlier (according to either rank) firm name generates advantages in foreign markets that speak more proximate (to either language) languages.

Cheng et al. hypothesise that firms with names earlier in the Roman alphabet will have an advantage in exporting to countries with greater language proximity to English, and firms with names earlier in the Chinese lexicography will have an advantage in exporting to countries with greater language proximity to Chinese. Using Chinese customs data from all export transactions over the period from 2004 to 2013, they find that:

...firms with lexicographically earlier character names export more to destination countries that have greater language proximities to Chinese, while firms with lexicographically earlier romanized names export more to destination countries that have greater language proximities to English. Such lexicographic biases are especially strong among exporters with low visibility in their businesses. Quantitatively, when an exporter moves down the lexicographic ranks from the 25th percentile to the 75th percentile, and meanwhile the foreign market's language proximity increases from the 25th percentile to the 75th percentile, its export volume decreases by one to 3%.

The effects are not huge, but they are statistically significant - there is 'alphabetic' discrimination for exporting firms. Why would that be the case? As the above quote notes, the effects are especially strong for firms with low visibility in their industry. Cheng et al. note that:

Lexicographic ranks supposedly matter more for obscure firms. The most established firms in a market are unlikely to be harmed by having an initial Y or Z in their names. A visual “first come, first served” seems to be a natural routine for importers who search obscure firms.

Do importers really search for exporting firms from an alphabetical list? While Cheng et al. do find some evidence that supports this, it would be interesting to see more support for this. In the meantime though, it appears that not only is Aaron A. Aardvark becoming a well-respected economist, but Aardvark Enterprises is becoming a top exporter (to English-speaking countries, at least).