VoxEU Column Environment International trade

The remarkable pattern of US water use

Water management is a major challenge today. To guide efficient water allocation, it is essential to understand the drivers of water use. This column sheds light on this issue using US data from the 1950s until today. The findings show that US water withdrawal has stabilised, and has even decreased in the past decades. Technological improvements have been crucial towards that end. However, the shifting demand from agriculture and manufacturing to less-water intensive sectors has been just as important.

With every new report of water crises in California, Australia, Spain, Brazil and other places, it becomes increasingly clear that adequate water management is one of the major challenges the world faces today. Contrary to discussions about climate change, and no doubt related to the tangible, local impact of water crises, the public has a more willing ear for the often alarming news about this essential resource.

To properly guide discussions about a more equitable and efficient water allocation, it is essential to understand the drivers of water use as well as the mechanisms behind increased water productivity, as Gleick (2003 a, b) points out. In this column, we lay out the facts about US water use, show what drives water productivity, and draw some lessons for the US and the world (Debaere and Kurzendoerfer 2015).

The facts

The US is one of the few countries in the world that has long-term, high-quality water withdrawal data to support an economic analysis of long-term water use. The US Geological Survey (USGS) just released the last data instalment. A few facts stand out:

  • US water withdrawals more than doubled between 1950 and 1980, but have remained constant – and even slightly decreased – in the last 30 years. 

The lowest line in Figure 1, which traces growth in US water use since 1950, shows this growth and subsequent tapering.

  • GDP growth in the US has decoupled from water use.

While the top line in Figure 1 which depicts GDP growth since 1950 has only risen, actual water use (the lowest line) plateaus.  

  • Increased water productivity has made the decoupling of GDP growth from water use possible.

The GDP in 2010 was 6.7 times what it was in 1950, whereas water use was less than twice its 1950 level. In sum, water productivity has more than tripled since 1950. The difference between the top and bottom line in Figure 1 displays the water productivity gains.

  • Agriculture and electricity generation are responsible for over 75% of total water use, and manufacturing and services account for less than 15% (see Figure 2).
  • In terms of water use per dollar of gross output, agriculture is far more water intensive than manufacturing and especially more water intensive than services.

Figure 1. Decomposition of US water use since 1950

Source: USGS data, own calculations
Notes: Actual water use tracks GDP growth, sectoral change, and technological change.

Figure 2. Water withdrawals by sector, 2010

Drivers of increased water productivity

To understand what drives US water productivity, we build on Leontief’s input-output analysis to link US GDP and the components of final demand to the US water data.

We first investigate the role of globalisation. Because of the persistent trade deficit of the US, one may wonder whether its ability to use less water is at the expense of the rest of the world. Is it the case that its water saving hinges upon importing more water-intensive goods? While the US importing more than it exports does help, we find that the trade deficit only accounts for 8.2% of all the water productivity gains. This finding is important and suggests that some of the US water savings could potentially be replicated in the rest of the world.

  • We find that technological improvements have been crucial in allowing the US to produce each dollar of its GDP with increasingly less water.

But even though technological progress is a large part of the story and is often the focus in the literature, it is not the only factor. 

  • The changing structure of the US economy, with demand shifting away from agriculture and manufacturing toward the far less water-intensive service sector, has affected water use in a big way.

The second-highest line in Figure 1 shows the hypothetical scenario of what water use would have been if one let economic growth and the sectoral composition of the US economy evolve at their actual rate–but holding technology fixed at 1950. The difference between the top line and the second-highest line then captures water savings due to sectoral shifts. As one can see, between 1950 and 2010, up to 50% of water productivity gains come from sectoral shifts. Note that the growing role of services in the US as well as in the world economy is good news, since services is by far the sector with the highest water productivity. The ongoing sectoral shift in the world economy suggests that water use will slow as the share of services in the global economy increases.

Technological improvements are responsible for the remainder of the water productivity gains–at least 50%. Since actual water is influenced by technological changes as well as structural shifts in the economy, the difference between the lowest line in Figure 1 (actual water use) and the second-highest one marks the contributions of technological improvements. The sizeable impact of technological improvement is a welcome finding if one considers technology an opportunity for replication abroad. Indeed, transferring technology can be a more actionable way of bringing about less water use, especially compared to the slow-moving process of structural shifts toward a less water-intensive service economy. What is striking is that two-thirds of the productivity gains are due to technological improvements in the electricity-generating sector. This finding is promising from a policy perspective. Water savings in the electricity-generating sector can be traced back to the Clean Water Act of the mid-1970s that encouraged, among other things, closed-loop cooling, which in turn encouraged water conservation.


Gleick, P H (2003a), “Global Freshwater Resources: Soft-Path Solutions for the 21st Century.” Science Vol. 302, 28 November, pp. 1524-1528.

Gleick, P H (2003b), “Water Use”, Annual Review of Environmental Resources, 28, p. 275-314.

Debaere, P and A Kurzendoerfer (2015), “Decomposing U.S. Water Use since 1950, Is the U.S. Experience Replicable?”, CEPR Discussion Paper 10573

2,835 Reads