Monday, January 25, 2021

What are the Australian and Spanish grape-growers doing?

Commenting on the effects of climate change in the wine industry, Jancis Robinson (What price terroir in an era of climate change?) recently commented:
Here then is the dilemma which will face Bordeaux châteaux in the next 20 to 50 years. Might they move to a cooler region and plant their present famous varieties, Cabernets Sauvignon and Franc and Merlot, or, perhaps stay put in Bordeaux and grow more heat-tolerant varieties such as Syrah and Grenache?
The latter option is being actively pursued in several parts of the world, notably Australia, where the climate is now notably drier and hotter than it was when I grew up there, last century.* Robinson continues:
For presently cooler regions, such as Burgundy, or Central Otago in New Zealand, it is still possible to choose more adapted varieties as warming occurs ... However, for regions already enjoying quality reputations with more heat-tolerant varieties, such as Australia’s Barossa Valley with Shiraz, the situation is much more difficult as there are few well-known varieties adapted to hotter climates.
https://www.etsy.com/listing/495670262/erte-1982-heat-spell-lady-on-grape-vine

There are clearly two topics here: what the Australians (and others) are already doing, and what the Europeans might start doing. In this regard, the Australians are trying Spanish, Portugese and Italian grape varieties, whether well-known or not, but the Spanish seem to be trying French varieties!

The evolution of Australian wine

Australians originally planted grape vines familiar to those growing them, from the Old World. However, through time, the vineyards ended up being dominated by the "usual suspects" — the well-known international varieties. However, it has now been noted that the wine grapes are ripening up to two days earlier each year (Australian wine under threat from climate change, as grapes ripen early).

So, there is now increased interest in what have become known as "alternative grape varieties". In one sense, these consist of whatever varieties seem worth trying in a changing climate, or even in response to interest from wine drinkers. However, in general, the varieties being trialed are the tried and true ones from dry climates in the Old World. These are, then, mostly those from close to the Mediterranean. This actually has a long history, with Coriole first importing and championing Italian varieties (such as Sangiovese and Fiano) back in 1985.

This topic has been the basis of a recent research articles (eg. Alternative varieties for the Australian wine industry to help winegrape producers in a changing environment and market. 2020. Wine & Viticulture Journal 35(3): 52-57). There are also online discussions, such as Alternative Australian varieties and Australian alternative varieties: the ones to watch, three years on. There is even an entire website devoted to the topic (Vinodiversity). Some of the varieties being actively pursued include: Assyrtiko, Fiano, Nebbiolo, Nero d’Avola, Grüner Veltliner, Tempranillo, Touriga Nacional, and Vermentino.

This is not to say that all of Australia is like this — it is a big continent, after all. In particular, Tasmania's cooler climate is still suited to those traditional grape varieties currently in strong demand — Pinot noir: 44% of the area; Chardonnay: 27%; Sauvignon blanc: 11%; Pinot gris: 8%; Riesling: 6% (Introduction to Tasmania’s wine and regions).

Finally, note that there are actually two potential driving forces behind the move to alternative varieties: a drying climate, and customer demand. If these both lead in the same direction, then all will be well in both the vineyard and the winery (Varietal diversity to drive demand). Otherwise, we might possibly see some future confusion within the Australian wine industry. This potentially occurs because of the New World propensity for naming wines after their component grapes, so that these grape varieties need to be known by the customer (recently discussed by Tom Maresca). In the Old World,  there are protected wine names, which are composed of sometimes quite un-namable varieties — for example, can you name the varieties that go into Rioja Blanco, from northern Spain?

Changing Spanish vineyards

In contrast to this, the Spanish grape-growers seem to have a very different idea. Recently, the American Association of Wine Economists (AAWE) put a couple of tables on their Facebook page:
   Red wine grape varieties in Spain: Vineyard area and age structure
   White wine grape varieties in Spain: Vineyard area and age structure
These data (originally from Estadísticas agrarias) show that relatively recent grape-vine plantings have included considerable areas of French varieties. Here are those with at least 1,000 hectares of vines:

Cabernet sauvignon
Syrah
Merlot
Chardonnay
Sauvignon blanc
Petit verdot
Pinot noir
20,535
20,155
13,044
7,674
5,087
1,866
1,063

It is difficult to believe that this has anything to do with improving the Spanish vineyards in response to ongoing changes in growing conditions. Indeed, it looks far more like an attempt to cash in one the retail popularity of certain French wines. If so, then given the quote that I started this post with, it looks doomed to failure, unless these are high-altitude or otherwise cooler vineyards.

Indeed, it may have much more to do with the amount of bulk wine that travels from Spain to France, to beef up the northern wines — we need to get the varieties right, after all. Things ripen much better in Spain. For example (The differences between high- and low-elevation wine):
Médoc, also known as the Left Bank of Bordeaux, is one of the lowest-lying wine regions in the world. This enabled grapes like Cabernet Sauvignon and Merlot to ripen in a marginal climate for these varieties before climate change. Higher altitudes would have been too cool, and even low-lying vineyards often struggled to ripen grapes fully.



* Those wildfires that you are increasingly hearing about are simply one obvious by-product (as well as deliberately lit ones). Trust me, I know. In my previous life as an environmental biologist (30 years ago) I used to study the effects of bushfires on Australian plants, and how those plants cope with what is obviously an extreme situation. Well, that situation is now much more extreme than it was back then. The same is true of California, which I first visited in 1990.

Monday, January 18, 2021

How soon is wine consumed after purchase?

I have always claimed that: if you have some nice food to eat, then you should have some nice wine to go with it; and if you don't have nice food, then you need some nice wine to make up for this. The main question, then, is where to get the nice wine, which is the only acceptable sort.

We have repeatedly been told over the past few decades something like this: "95% of wine is drunk within 24 hours of purchase" (eg. Cork vs. twist off: we have been screwed;   The perfect wine, on hand). But where does this "statistic" come from? No-one ever seems to quote a source — it is apparently just one of those things that "everyone knows". Indeed, I have included at the bottom of this post a selection of a few other similar statements from around the web.


To me, this claim looks like a factoid or urban legend. After all, the actual number seems to be quite non-believable. For example, people buy bag-in-box wine in order to have some open wine on hand whenever they feel like a glass, not to binge drink the entire container within one day; and this packaging type accounts for an awful lot of wine purchases. Similarly, a lot of the bottled stuff is bought by those people with a genuine "wine interest", and they are not likely to chug-a-lug the stuff it as soon as they get it home — how we are we supposed to talk about our wine collection if we have just drunk it?

Note, also, that this claim is a very different thing from saying that most wine is produced with the intention that it will be drunk soon after production. There seems to be no obvious objection to the suggestion that most wines released are meant to be drunk within 1—3 years of bottling, while they are young and fresh, especially white and rosé wines. This refers to time from production, not time from purchase. For example, I usually drink rosé wine of the current vintage, although I did recently find one that had been sitting in the store for a year — Minuty Prestige 2019 (Côtes de Provence), and very nice it was, too.

Now, I am not going to spend my time tracing the origin of the magical "95" number (or "94", "90", etc). There are plenty of people on Youtube, and similar social media, who do that sort of thing (eg. LEMMiNO;   Cheddar;   Today I Found Out). Of far more interest to me, at least, is: what is the real number? After all, there must be one, although it is unlikely to be the same number in each wine-drinking culture around the globe.

The obvious place to look is the USA, since Americans seem to spend a lot of time and resources studying each other. To this end, Liz Thach & Angelo Camillo conducted a survey of 1,200 American wine consumers back in 2018 (A snapshot of the American wine consumer in 2018). One of the questions was: "When you buy a bottle of wine, how soon do you USUALLY drink it?"


The above graph shows you the compiled answers. Now that looks more like a number I am prepared to believe. Like all consumer surveys, we cannot treat the number as being very precise, since it depends on the sampling strategy used to gather the respondents; but there is no reason to think that this number is not in the right ballpark. That is, only one-fifth of people consume wine within a day of purchase, and another one-third within the next few days. Very sensible of them, I am sure.

On the other hand, only one-tenth of the people seem to consider cellaring their wines. These people are even more sensible, of course.

If we are not going to drink our wine within a few days of purchase, then this leads to the topic that I have previously looked into: How long should we cellar our wines? Apart from the obvious answer ("until they are ready"), we do need to consider whether we are buying an already mature wine. We might, then, let the bottle settle for only a few days, to allow the contents to recover from any possible travel shock, even though we wish to consume it sooner rather than later.



Does the price of a bottle of wine go up in price each year?
Something like 98% of wine is drunk within 48 hours of purchase and is meant to be consumed young.
Six things you probably don’t know about wine
As much as 90 percent of the wine bought in the U.S. is drunk within 24 hours ... And, depending on the study, some 95 percent of all wine purchased in the U.S. is consumed within a week.
Wine picks for the starter collector
the vast majority of wine in America — estimates range from 70% to 90% — is consumed within 24 hours of purchase
Do you drink it now or drink it later — That is the question!
Years ago, I heard that something like 50% of wine is consumed within 48 hours of purchase. I can’t verify that statistic, but I do know that most of the wine we sell gets enjoyed promptly!
80% of all wines bought in the US are consumed w/in 48 hours
Apparently this is straight from the LCBO [Liquor Control Board of Ontario]: 90% of purchases at the LCBO are consumed within 24 hours, and most people don’t know what they want to buy before they enter the store. I generally have a corkscrew in my hand at checkout, so on average, I'd say that I wait about 6.75 seconds before opening a bottle I purchase.

Monday, January 11, 2021

Wine prices 2000-2020

In my previous post I looked at 20 years of buying Australian wine in Australia (1980-2000). I noted an exponential increase in the prices of the wines, along with only a linear increase in prices due to inflation. This neatly complemented recent media comments about "prices for the established trophy wines [having] skyrocketed in recent years" (Beware the hype) — it has been happening to ordinary wines, too.

I moved to Sweden at the beginning of this century, but I continued buying Australian wine, so that I can directly compare the most recent 20 years with the previous 20. Interestingly, in this post we get to see something really quite different in terms of price, compared to last time. Wine buying is not as simple as you think!


Trying to get Australian wine in Sweden is obviously far more difficult than getting it in Australia. There is certainly a selection available, but it is relatively limited. So, at times I have also bought wine in other European Union (EU) countries, to supplement the supply. The best source has been the United Kingdom, for historical reasons; and it is still the biggest European importer of bulk and bottled Australian wine to this day (see Top bulk wine suppliers to the UK, 2019 v 2020). So, I have a reasonable number of wines to study.

Indeed, the first graph below shows that I have purchased in Europe 570 Australian wines, from 154 wineries, compared to 581 wines (from 118 wineries) back in Australia; so, I have kept pace. As you can see, I first started getting wines from the UK in 2007 (hence the peak that year), and stopped in 2018. I started getting wines from elsewhere in 2008, either on holiday travel or by delivery; my last purchase from Germany was in 2018. I also got some wines on eBay, from 2011 onwards. The big peak in 2016 involved getting a few mixed dozens that became available locally.

Number of Australian wines purchased

Overall, 30% of the Australian wines were not purchased in Sweden. Obviously, I was targeting selected wines that were not available locally. So, it is not surprising that I paid more for those wines than for the local wines — an average of 216 SEK versus 169 SEK (ie. 25% more). The currency exchange rate has varied a lot over the past 20 years, but at the moment US$ 1 ≈ 8½ crowns, so we are talking about wines in the vicinity of US$ 20, here. Some of the price difference is also due to transport costs.*

In my last post, for my analysis I used those 453 wines that had both vintage dates and prices in my (hand-scrawled) records — most of the wines were the current vintage release, so that the vintage year is closely related to the purchase year. This time, there are 522 such wines in my (much more neat) spreadsheet. These are plotted in the second graph. The horizontal axis shows the vintage year of the wine, while the vertical axis indicates the Swedish crown price at the time of purchase. Each point represents one wine, with the pink points being those purchased outside Sweden. The latter non-Sweden wines were converted to SEK (from £ or €) using the exchange rate at the time of purchase.

Australian wine prices by vintage date

Clearly, this graph does not illustrate an increase in wine price through time. Indeed, the exponential price model that I illustrated in the previous post (and which is common in economics) accounts for only 2% of the variation in price — this means that the average purchase price remained pretty constant across the 20 years.

This is not to say that the actual wines remained constant in price, since there is an expected increase in price due to inflation. Even non-necessity goods, like wine, do not remain constant. This increase is shown as the red line in the graph. The inflation data that I used comes from Sweden Inflation Rate 1960-2020. I started with an average wine price of 150 SEK for the 1997 vintage, and increased this each year by the annual inflation rate. The inflation data show a roughly linear increase in price over the two decades, amounting to 30%.

So, what is going on here? It seems to me that this is likely to be an example of an issue recently discussed by Huon Hooke (Price limits with wine):
The truth is, we all have our price-limits, whether or not we stick to a strict budget. But it’s always puzzled me that some people’s price limit for wine never seems to increase. 
I’ve known people who have been buying AUD $15 wine for decades. And they grizzle that it’s getting hard to find decent AUD $15 wine. Well, stop being a tight-wad, is my response, usually muttered under my breath. There is this thing called inflation. You don’t expect your other living expenses to stay the same for years, do you? So why wine?
Well, it seems to me that I have been caught out doing exactly this. I should be willing to pay 30% more for my Australian wine than I did 20 years ago, and yet I have not increased my wine "price limit" one iota. I am still targeting the same price range as when I moved to Sweden at the turn of the century. Mind you, I am still finding good quality wine in that price range (US$ 15-25), so I may not change my behavior any time soon. Moreover, I am a retiree, so I am unlikely to increase my expenditure any time before winning the lottery.

We have been told that Price affects the way people experience wine. That is, given exactly the same wine, but labeled with two different prices, we will mostly prefer the apparently more expensive wine. This being so, I keep away from the US$ 10 wines, because I don't want my brain to think that my wine is awful; but on the other hand, I also keep away from the US$ 50 wines, because I decline to pay that much for flavored alcohol.

Australian wine prices by purchase date

As noted above, the vintage year is closely related to the purchase year for the wines, except in those cases where I specifically bought back vintages. To check whether this has an effect on my analysis, the third graph plots the wine prices (vertically) along with the date of purchase (horizontally). This graph includes all 570 wines, rather than just the 522 plotted in the graph above.

This graph shows little alternative information, except that there are nine wines purchased in 2016 that exceed 400 SEK — these are part of a collection of back vintages bought from the UK, plus one on eBay.

Conclusion

For the first two decades of my Australian-wine purchasing activities, the bottle prices went up exponentially, while for the next two decades they remained constant. I do not conclude from this that wine prices stopped increasing. Instead, I later adjusted my behavior in order to maintain my wine-price limit. For example, my analysis of the data in the previous post indicates that I was paying an average of 184 SEK per wine bottle at the end of the first 20 years, but my analysis in this post indicates that I paid an average of only 169 SEK across the second 20 years (8% less). What a cheapskate!

It is, of course, somewhat more difficult to maintain this rate these days.



* One method for getting wine from the UK involved my wife attending EU meetings in London, which she did as part of her job. I would get some wines sent to her hotel, and she brought them back to Sweden in her luggage, thus saving on the (often quite serious) transport costs. This would net me six bottles every couple of months. This all stopped when the UK first announced Brexit, and the EU meetings were immediately moved to Amsterdam. The UK is still struggling to implement Brexit (including having asked for time extensions), but the EU has been ready for more than 2 years!

Monday, January 4, 2021

Exponential increase in wine prices (1980-2000)

Recently, both Neal Hulkower (The cost of drinking wine history) and Eric Asimov (How income inequality has erased your chance to drink the great wines) have commented on the prices of modern high-quality wines, compared to The Good Old Days. Well, I am just as old as they are, so I figure I can do this, too.

Neal's commentary was based on a mathematical analysis of his own records for 450 wines that he had tasted between 1969 and 1979 (What can I still afford to drink? AAWE Working Paper 254). I wrote a blog post about this, plotting his data as graphs (The outrageous prices of modern high-quality wines).


Well, The Good Old Days (last century) for me were not quite the same as for Neal (or Eric). We all have our price limit, and that applies to flavored alcohol, just as much as anything else. In Australia back then, we did not drink much "foreign" wine, because we could not afford it (as I noted in Tasting great wines?). So, if I am going to look at the wines I bought way back when, then the data will pertain to Australian wines, not American or European. (A note on a couple of the actual wines is at the bottom of this post.)

This is not to say that I drank no imported wines, because my records for 1982-2001 show 76 such wines, from 61 wineries. However, the Australian wines total 581, from 118 wineries, which is a much better sample size. So, the data discussed below refer only to the local wines. Here, I will look at the wines from the first 20 years of my time as a wine drinker (see Some personal anecdotes), and I will discuss the most recent 20 years in the next post.

Of these early wines, 453 have both vintage dates and prices in my (hand-scrawled) records; and these are plotted in the first graph below. The horizontal axis shows the vintage year of the wine, while the vertical axis indicates the Australian dollar price at the time of purchase, with each point representing one wine.

Most of the wines after 1979 were the current vintage release, so that the vintage year is closely related to the purchase year. These are shown by blue points in the graph (one per wine). The green points are all back vintages that were still available, if you were prepared to spend some time scavenging in the liquor stores. In most cases, I purchased only one bottle of each wine, on the principle that tasting a wider range of wines was more interesting and educational, rather than focusing on a few favorites.

Australian wine prices up to 2000

There are several features of the graph that we could focus on here. First, the post-1979 wines increase in value exponentially, as found by most previous commentators on wine prices. This is illustrated by the black line in the graph. This regression line accounts for nearly 50% of the variation in price, which is really quite strong, in the field of economics.

Some of this increase may, of course, be due to increased vinous ambition on my part, or even an increase in my salary, leading to me purchasing better and better wines. However, I do not think that these are large effects, as I have always kept a pretty tight reign on the price that I am prepared to pay (the affordability limit, as it is called).

So, what we are looking at here is mostly an increase in the price of Australian wines over a 20-year period. Even wine of reasonable quality became much less affordable during those 20 years, just as has been observed by Neal Hulkower and Eric Asimov for high-quality non-Australian wines.

A second point, then, would be to look at the expected increase in price due to inflation. We do not expect our living expenses to remain constant over the years, and the same applies to wine, even though it is one of the non-necessity goods. The inflation data that I have used comes from Australian Inflation Rates: 1949 to 2020. I started with the regression estimate of the average wine price for 1979, and increased this price each year by the annual inflation rate. The result is shown as the red line in the graph.

Note that the wine prices kept pace with inflation for the first decade of the data; but from the early 1990s onward the prices increasingly outstrip inflation. The inflation price increased roughly linearly with time, while the actual wine prices went up exponentially. Early on, there is thus little difference between the two, but not later on. So, maybe this provides a neat definition of The Good Old Days for my generation — when wine price increases kept pace with inflation but not more.

Increase in the price of wines bought twice

A third thing we could look at is the difference in price when I bought the same wine at different times, often several years apart. My records show 25 such occasions; and these are plotted in the second graph. Here, the first purchase price is shown horizontally and the second price vertically, with each point representing one wine. The pink line shows equality between the two prices.

Not unexpectedly, in most cases the wine cost more the second time, although occasionally bargains could be found. More interestingly, though, the prices here also increase exponentially, as illustrated by the black line in the graph. This regression line accounts for 70% of the variation in price. So, it will cost you dearly, the longer you leave it to purchase your bottle of wine.

Conclusion

Wine prices increased exponentially in Australia between 1980 and 2000, while inflation increased wine prices only linearly. Therefore, wine became increasingly unaffordable during that time.

In spite of this, wine itself became much more popular among a broader range of people, at the expense of the standard cliché image of Australians swilling beer. One cannot really begrudge this — the standard of the wine increased enormously, along with the price. Cheap fortified and sparkling wines were the fare of my parents' generation, but not of mine (The evolution of Australian wine). The Good Old Days for me were better than for my parents (winewise, that is).



The first top-of-the-line bottle I ever bought was the Penfolds 1972 Grange Hermitage (as it was then called). This particular wine was released relatively cheaply. The 1971 vintage was much-lauded (Penfolds Grange — Best vintages, prices, history), and the 1972 was thereby over-shadowed. At the same time, Penfolds needed some cash flow, so they under-priced the 1972. This was still available (if you knew where to look) in 1983 for a mere A$25, which would be A$90 today allowing for inflation. The current Penfolds Grange Bin 95 retails for ten times this amount, of course. The 1972 can still be had for A$1,000, or so.

The first wine I bought from the year of my birth was the Saltram 1958 Bin 20 Claret (Museum Release). This is a wine most of you have never heard of; but you can still read about it, because it was one of the better wines from a legendary winemaker (Peter Lehmann’s Barossa Valley masterpieces). I paid A$30 back in 1990, which would be A$65 today — an auction bottle sold in 2008 for A$289 (Langton's Top 500 List 2008).