Insights from the Washington Vineyard Acreage Report

The recently released Washington Vineyard Acreage Report, administered by the National Agricultural Statistics Service (NASS), provides detailed estimates of Washington State vineyard acreage as of January 1, 2017. The last report was published in 2011. In this entry, I discuss some of the highlights from the report, its limitations, and implications for the Washington wine industry and grape market.

The state added 11,596 acres between 2011 and 2017 – an average of 1,933 acres per year. The pace of growth has moderated relative to the prior five-year period when acreage increased by 2,570 per year. Since 2011, the number of planted vines has expanded at a slightly faster rate than planted acres due to an increase the density of plantings.

Washington now has 55,445 acres of wine grapes, including both bearing and non-bearing acreage. While the total still pales in comparison to California’s 600,000+ acres, Washington has grown at more than twice the rate since 2010, and its share of West Coast plantings increased from 7.3% to 8.1%.

Growth has been relatively wide-spread as all but two AVAs, Columbia Gorge and Puget Sound, recorded gains in acreage. The Horse Heaven Hills AVA added the most acreage; 4,325, while Red Mountain grew at the fastest rate; 48%.

Note: the Columbia Valley total in the above chart includes the Ancient Lakes, Lewis-Clark Valley, and Naches Heights AVAs, as well as all areas outside of AVAs entirely, as they can't be distinguished separately.

The report indicates considerable churn in varietals within Washington State, as growers continue to tweak plantings to site specific conditions and respond to changes in market demand. The long-standing shift toward red varietals shows no signs of abating, as acreage devoted to reds grew by 43% and now represents 65% of the state total. Conversely, white varietals increased by just 4%. Red grapes now account for the majority of acreage, and grew at a faster rate than white varietals, in all major AVAs except the Yakima Valley.

After decades of contention, Cabernet Sauvignon has clearly become Washington’s signature varietal. Cab accounted for 70% of the state’s acreage growth between 2011 and 2017, and now represents 34% of the total, which compares to less than 20% in California.

• Cabernet Sauvignon plantings expanded by 8,315 acres to a total of 18,608, an increase of 81%.
• Cabernet now represents 60% of planted acreage in the Red Mountain AVA, which exceeds the grape's share in Napa Valley.
• While Merlot is still the state’s second most important varietal in terms of acreage, plantings increased by just 10%.
• Syrah continued its rapid rise in Washington – it added nearly 1,500 acres – a 47% gain.
• The state’s leading white varietals, Chardonnay and Riesling, expanded by just 2% and 6% respectively. 

The survey recorded a total of 50 wine-grape varietals growing in Washington, but the top five alone account for 84% of the state’s total acreage, up from 81% in 2011. Growth estimates varied widely among the 19 secondary varietals that are singled out in the report: 7 gained acreage and 12 lost. Surprisingly, the survey recorded declines for several key Rhone and Bordeaux red varietals that have been rising in stature, although these estimates should be viewed with caution as explained in the next section.

• Malbec and Pinot Noir acreage, while still insignificant in absolute terms, both expanded by more than 50%.
• Cabernet Franc, Petit Verdot, Grenache and Mourvedre acreage declined by between 15% and 30%.
• Pinot Gris, the state’s third ranked white grape by acreage, led all whites with an increase of 547 acres.
• Sauvignon Blanc is also playing a more prominent role in Washington, as plantings grew at a robust rate of 24%.
• Conversely, Chenin Blanc acreage declined by 68%, and Gewurztraminer fell by 48%.

Statistics should always be viewed with a critical eye – this is particularly true for wine industry data. The Vineyard Acreage Report is based on a voluntary survey, so the figures it provides are estimates – not facts – as even well-designed surveys are subject to error. I’d like to highlight several important limitations that users of the report should be aware of:

1. Low response rate: NASS received responses representing just 52% of the total acreage it identified (compared to 87% in 2011), and some contained only partial information. Based on the responses it received, NASS used an algorithm to derive estimates for the state, AVAs, and varietals. If the responding acreage was highly representative of the state’s total acreage this would not be problematic, although the estimates would still be subject to sampling error. But, the survey was not based on a random sample, rather respondents “self-selected,” so there is no reason to believe the responses are representative, and no way to quantify margin of errors. 
2. Measurement error: Because the survey is voluntary, there is no guarantee that respondents took the time to confirm their responses with actual vineyard records. Misinterpretation may also be an issue. For example, the question pertaining to year planted might be interpreted as the year when the rootstock was planted, or when the top varietal was grafted.
3. Data suppression: Finally, a good bit of the data has been deliberately suppressed to preserve the confidentiality of respondents. For example, no totals are given for the Ancient Lakes and Lewis-Clark Valley AVAs for this reason. Unfortunately, when an entry is missing there is no way to tell whether it represents a “zero” value or has been suppressed - which complicates interpretation.

There are anomalies that suggest at least some of the reported figures are not reliable. For example, the 2017 report indicates that 14,385 acres were planted prior to 1996 – but the 2011 report found only 11,962! The report also suggests that Cab Franc acreage declined by 30%, yet the Grape Production Reports covering the same period indicate that Cab Franc output grew by 72%. Lower yields and coverage differences may explain part of the discrepancy, but anecdotal evidence supports the hypothesis that Cab Franc plantings are expanding in Washington, not contracting.

Overall, I see nothing to suggest that NASS did anything other than an exemplary job with available resources. Nonetheless, due to the limitations discussed above I believe margins of error are likely to be high for some components of the report. My conclusions are as follows:

• The state-wide total acreage figure is likely to be reasonably accurate.
• The state-wide total acreage figures by varietal are likely to be fairly accurate for primary varietals – but less so for secondary varietals.
• The total acreage figures for large AVAs are also likely to be fairly accurate – but those for smaller AVAs with more fragmented ownership are likely to be less reliable.
• The acreage figures by varietal for individual AVAs are likely to be less accurate in general, particularly those for secondary varietals in smaller AVAs.
• The vine totals are likely to be less accurate than the acreage totals.
• The year planted figures should be viewed with skepticism.
• Comparisons between the 2011 and 2017 reports should be undertaken with caution as they are based on different surveys.

Given the limitations discussed above, as well as the fact that the report does not distinguish between bearing and non-bearing acreage, the estimates are clearly insufficient for deriving precise forecasts of potential future grape supply. However, it is possible to at least draw some broad-brush conclusions.

The acreage estimates by year planted suggest there was a modest decline in plantings in the 2014 to 2016 period relative to the prior two years. Although the margin of error for these estimates are likely to be high, the figures are not inconsistent with anecdotal evidence that suggests there hasn’t been a major shift in planting trends in either direction. Given this, it would reasonable to expect "potential" grape output to expand by around 3% to 5% per year over the next several years, as vines planted between 2014 and 2016 continue to mature. Provided the weather cooperates, this should create an ample supply of grapes for Washington to continue to grow its market share, as California’s vineyard acreage appears to have peaked, at least temporarily.

The Washington grape market has been in a relatively balanced state in recent years as steady growth in demand for Washington wine has kept pace with the increasing supply of grapes. This stands to reason as new acreage has generally been contracted by wineries in advance, while speculative plantings have been the exception. However, I anticipate some challenges on the demand side of the equation going forward.

Recent metrics from multiple sources indicate that growth in U.S. wine sales by both value and volume decelerated in 2017. My own analysis of tax reporting data from the Washington State Liquor and Cannabis Board suggests that Washington wine sales by volume actually contracted in 2017 (I’ll be publishing definitive numbers soon). And, according to TTB data, Washington wine inventories increased in 2017, which is not surprising given the slowdown in demand and large 2016 harvest.

Unless this trend reverses course or yields come in below normal, there is a growing risk that grape supply will outstrip demand over the next several years. Of course, this conclusion is based on a simple back of the envelope analysis - a formal analysis would be needed to confirm it as well as to estimate the size of any potential imbalance.

Given uncertainty in the outlook for demand growth, growers should be cautious and strategic in their planting decisions. The grape market is not monolithic, and even if supply does outstrip demand in the aggregate, some segments - in terms of varietals, AVAs, and quality tiers - will inevitably be better positioned than others going forward. A well-informed and appropriately segmented market analysis can provide a sound foundation upon which to both define the risks and identify the opportunities.

I’d like to close by encouraging all growers to respond to the Vineyard Acreage Survey in the future. It is the most comprehensive source of wine-grape plantings data for Washington State – and a higher response rate will result in more reliable metrics – which are the key to informed decision making.