Monday, December 9, 2019

When will world wine consumption finally catch up with production?

There has been much recent talk about world wine production, not least in relation to global climate change. Among all of this, we have been told, by the International Organisation of Vine and Wine (OIV), who collate worldwide records, that the 2019 production level has returned to near “normal”, after a small grape crop in 2017 was followed by a bumper crop in 2018.

Furthermore, concern has been expressed that the 2018 wine production, much of which is now being released to the consumers, has produced a market glut. This is obviously not good for the long-term viability of wine producers, so it is worthwhile for us to look at current trends in both the worldwide production and consumption of wine.

Production

The data graphed below come from the OIV database, which has collected worldwide data since 1995 (ie. the past 25 years). The first graph shows the total level of wine production from those countries with commercial production that exceeds a specified minimum. The data for 2018 is still provisional, since not all producers have yet been reported their final numbers; and the 2019 data are preliminary, as wine production is still ongoing.

World wine production 1995-2019

As you can see, wine production varies considerably from year to year, but there is no consistent trend upwards or downwards, nor are there any repeated cycles. So, we may conclude that current wine production is simply varying stochastically around a particular “average” level (currently c. 270 million hectoliters annually) — this is indicated by the dashed line on the graph.

The 2017 production was the smallest since the beginning of the 1960s, while the 2018 production was exceeded only by that of 2004 (for the longer-term trends, see my post The smallest global wine production for 55 years). The estimated 2019 production level, on the other hand, is only c. 3% less than the 25-year average. We therefore cannot claim that this year is in any way unusual, in terms of production — it was the previous two years that were unusual. Note that world production is dominated by European producers, who contribute about two-thirds of the wine (62-69% annually during the current century).

The consequences

The importance of the return to normality in 2019 results from the obvious over-production of wine from the 2018 grape harvest: “with demand continuing to soften in volume terms, this reduction in supply comes as welcome news for wine producers” (Global wine supply moves closer to demand in 2019). In Europe, for example, there are reports of “a fairly well-inventoried bulk wine market in Italy for Prosecco, Pinot Grigio and Primitivo, as well as built up inventories of France’s Languedoc reds” (The ups and downs of the 2019 global bulk wine market).

In the USA, Napa grape prices for 2019 are well down, making it “a year where it’s tough to sell grapes and bulk wine, because most of the buyers already have sizable inventories because of 2018” (A historic harvest and a changing market: Napa's growers navigate grape glut). Furthermore, “the slackened thirst for purchases of excess North Coast grapes and wine is expected to extend into 2020, as vintners move the bounty of the 2018 harvest through sales channels that aren’t growing as fast as the influx” (As North Coast vintners move through ample wine supply, Napa County adjusts rules for small vintners).

One suggested response to this situation in Napa is for “vineyard owners strategically replanting some acreage this year, [thus] delaying future fruit production to correspond with market rebound” (Napa's growers navigate grape glut). This option has also been suggested in Italy (Barolo consorzio bans new plantings for three years to avoid wine glut).

Consumption

These are pretty serious suggestions, which obviously leads us on to a comparison of production with global wine consumption. The OIV data are shown in the next graph, with the consumption shown in pink.

World wine production and consumption 1995-2017

Clearly, there have been long-term trends in the consumption data — there was an increase in consumption until 2005, followed by a dramatic drop, returning to the 1996 level. The drop was then reversed, with the previous (2005) consumption level recovering by 2014 (9 years later). Wine consumption has shown a roughly linear increase since 2007.

Thus, we may conclude that production has remained static while consumption has very recently been increasing. We may then ask the obvious question: when will the level of consumption catch the level of production?

We can answer this question using a very simple linear forecast, as shown by the two dashed lines on the above graph. It is important to note that forecasts, by definition, assume that the current trends will continue unabated into the future, which may be rather naïve. Indeed, there have been media predictions that future wine consumption will actually plateau, or even decrease (eg. Worldwide alcohol consumption declines 1.6%).

Anyway, the forecasting lines for production and consumption intersect in the year 2034, which is 15 years away. Until then, production will continue to exceed consumption, and the risk of a wine glut will remain real.

As an aside, if we produce the same type of forecast based solely on the consumption data from 1995 to 2005 (when there was a also a linear increase), the forecast year of equality between consumption and production is 2029. Thus, the dip in consumption from 2005-2014 had the effect of delaying the forecast by 5 years.

An alternative forecast for future consumption is shown in the final graph, where consumption is treated as actually having already reached a plateau (at an average of c. 238 million hectoliters annually), with a dip in consumption during 2005-2014.

World wine production and consumption 1995-2017

For this forecast, both wine consumption and production increase only very slowly. In this scenario, consumption stays at <90% of production for the foreseeable future (the two lines slowly reach equality in the year 3555!).

Obviously, some of the pundits are concerned that this second forecast is more likely than the first one.