Knowledge and Use of Contracts, Digital Platforms, and Blockchain Technology by Small Scale Specialty Crop Growers: Focus Group Findings

In collaboration with AgLaunch and its Farmer Network, we conducted an online focus group with four small-scale, specialty-crop producers. Insights from the discussion may provide helpful information for Extension educators regarding farmers’ views and use of marketing contracts/agreements and digital trading, marketing, and payment platforms, including their experience trialing the features of different platforms and the advantages and disadvantages of the platforms used. The discussion also included farmers’ preferences for marketing their agricultural products, the effects of the COVID-19 pandemic on their operations, and farmers’ views and knowledge of blockchain technology, internet access, and mobile applications.

Here are the main findings of the focus group:

  • Participating farmers were small-scale (1 to 6 acres), specialty-crop growers of leafy greens, root crops, and other vegetables. Current certifications included Certified Naturally Grown (CNG) and USDA Organic, while past certifications included Good Agricultural Practices (GAP). The presence of a buyer requiring GAP certification drove the farmers’ decision to get certified.
  • Some digital trading/marketing platforms they use include Shopify and Barn2Door. Platforms they previously used or tested include Farmigo and Harvie for Community Supported Agriculture (CSA), Squarespace, and Local Line. They also discussed digital platforms to process payments at the farmers’ market, including PayPal, Square, Cash App, and (PayPal-owned) Venmo, as well as the pros and cons of these platforms based on their experience.
  • Their most important concerns regarding digital trading/marketing platforms and the features they seek in such platforms include the benefits outweighing the costs, convenience for their customers, ability to communicate with their customers, privacy and security, cybersecurity threats, customer service/support by the platform developers, training and technical assistance opportunities, and the platform’s ability to work/connect with other systems used on the farm (in other words, interoperability).
  • They agreed that the importance of communicating farm and product attributes to buyers through a trading/marketing platform depends on what their buyers want, their location, and their relationship with them. They also agreed that different buyers place different emphasis on these attributes.
  • They observed the repercussions of the COVID-19 pandemic regarding the closure of many restaurants and food-service institutions and the negative impact those closures had on produce suppliers. Some farms emphasized online sales, while others downsized or closed permanently.
  • They use or had used verbal or informal agreements instead of written or formal contracts. Key contract attributes include quality and processing specifications, payment method and turnaround, delivery and packaging requirements, delivery time, a guaranteed purchase, a guaranteed minimum price, and penalties for non-fulfillment. If formal contracts were an option, these would be important if they were selling higher volumes to wholesalers, selling to buyers with whom they did not have an established trust relationship, or if the transaction involved custom product requests for which they had to considerably invest additional time and resources. Depending on the parties’ risk tolerance, the higher their perceived risk from these transactions, the more they would favor formal contracts.
  • The farmers had heard about blockchain before, but their level of familiarity ranged from minimal to very familiar. After learning more about blockchain and smart contracts, most farmers expressed that they would feel comfortable using blockchain ledgers to store their information but thought the technology might be better suited for transactions with institutional buyers.
  • Their concerns about blockchain applications in agriculture and food supply chains included a need for a better understanding of these applications and their problem-solving capabilities, the cost and transaction fees, and long-term sustainability.
  • They expressed comfort using mobile applications in general. However, all farmers viewed internet access as an issue for rural and urban locations.

To access an Extension publication with more information on the experiences shared by farmers, please click HERE.

Did Pumpkins Cost You More in October 2020 Than in Previous Years?

It depends on whether you will be carving a jack-o’-lantern, decorating, or baking a pumpkin pie. And if carving a jack-o’-lantern, on what size. The U.S. Agricultural Marketing Service (AMS) surveys large retailers, and every week, publishes a national retail report of advertised specialty crop prices ending during a given period, usually an almost 2-week window. In this post, I take a closer look at the retail pumpkin prices for the Southeastern United States (i.e., Alabama, Florida, Georgia, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, and West Virginia) published by the AMS in the specialty crop retail reports released in October 2018, 2019, and 2020.

On average, advertised retail prices for all pumpkins in the Southeastern United States in October 2020 were 8.9% higher than those in October 2019, and prices in October 2019 were 8.3% lower than those in October 2018, leading to similar average prices in October 2020 to those in October 2018 (Figure 1). Average retail prices for extra-large pumpkins in October 2020 were also close to those in previous Octobers, without much fluctuation in prices throughout the month this year (Figure 2). However, it seems that average retail prices for large and medium pumpkins this October were higher than in previous years, with medium pumpkins seeing yearly price increases. While average retail prices for large pumpkins in October 2019 were 1.9% lower than those in October 2018, they increased by 7.9% from October 2019 to October 2020 (Figure 3). Average retail prices for medium pumpkins in October 2019 were 1.7% higher than those in October 2018 and 18.0% higher in October 2020 than those in October 2019. Data collected from 311 stores show that advertised prices for medium pumpkins were highest for the period ending during the 2nd-3rd week of October 2020, when unit prices ranged from $4.00 to $8.00 and averaged $6.83 (Figure 4).

 

 

 

Many factors can drive up the prices of consumer goods, like supply chain disruptions, government policies that boost demand and economic growth to the point that demand exceeds production capacity, and people’s or firms’ expectations of higher prices. Food inflation measures how much more expensive food has become over a certain period. According to the latest inflation data available from the U.S. Department of Labor Statistics, the average price of food at home in the United States rose 1.0% from October 2018 to October 2019 and 4.0% from October 2019 to October 2020. Thus, the increases in retail prices for large and medium pumpkins in October 2020 relative to October 2019 (7.9% and 18.0% for large and medium pumpkins, respectively) were higher than the food inflation rate of 4.0% for October 2020. Although the food inflation rates published by the U.S. Department of Labor Statistics capture many factors influencing food prices during the previous 12 months, this comparison might suggest that the cost of those two pumpkin sizes in October 2020 grew at a faster pace than recent estimates of the overall cost of food.

In contrast with other sizes, miniature and pie type pumpkin prices were lower this year than in previous years. On average, advertised retail prices for miniature pumpkins in October 2020 were 13.3% higher than those in October 2019, and prices in October 2019 were 17.4% lower than those in October 2018, leading to lower average prices in October 2020 than those in October 2018 (Figure 5). Average retail prices for pie type pumpkins in October 2020 were also lower than those in previous Octobers, with pie type pumpkins having yearly price decreases. Average retail prices for pie type pumpkins in October 2019 were 6.7% lower than those in October 2018 and 5.0% lower in October 2020 than those in October 2019 (Figure 6).

 

 

Whether you will be carving a jack-o’-lantern, decorating, or baking a pie, the figures in this blog post can give you more insights on how pumpkin prices at your local store compare with average prices at major retailers in the U.S. southeast region—and on whether those local price tags are scary or not.

 

What Can Google Searches Tell Us About Changes in Consumer Behavior Toward Food and Plants Beyond COVID-19?

Google Trends is an online tool that allows you to analyze the popularity of Google Search inquiries. If you enter a search term, the site randomly samples Google Search’s massive database to produce indicators of Google users’ past interest in that term. For a given region and time range, the site calculates an index between 0 and 100 for each point in time and produces a chart of interest over time, among other indicators. If analyzing a single search term, a peak value of 100 indicates the point at which the term was most popular for the region and time range selected (that is, the point with the highest percentage of searches for that term out of all searches conducted). The other values represent search interest relative to the highest point on the chart. So, if point A has a value of 100 and point B of 50, the term was half as popular at point B than it was at point A.

I used this tool last year for a talk at the Southern Agricultural Economics Association in which I illustrated how people in the United States, including consumers, lobbyists, and policymakers, are actively searching online for information on key topics related to the food industry (such as the terms local food, food waste, food miles, cottage food, pasture-raised, cage-free, vegan, GMO, bioengineered, or gene editing), and how sometimes their interest starts picking up years before legislators pass major food bills into law. Recently, Dr. Tim Woods at the University of Kentucky asked me what one of my graphs would look like if I included observations from the COVID-19 era, a great question that I will try to answer in this post!

Figure 1 below shows the average Google Trends scores for the terms Local Food, Food Waste, Cottage Food, Online Groceries, and Home Gardening from 2004 to September 2020 so far. Figure 2 shows the Google Trends scores for the same search terms but disaggregated monthly from January 2018 to September 2020 so far. Because I am interested in search interest for these industry segments relative to themselves, which in this case makes any spikes more meaningful, I first entered each term in Google Trends and then combined the data, instead of comparing the different terms against each other.

 

 

 

Here are a couple of things that these figures might tell us: Figure 1 suggests that search interest in the terms Local Food, Cottage Food, Online Groceries, and Home Gardening in the United States has increased during 2020, with Online Groceries showing the most noticeable change relative to previous years, and that search interest in Food Waste has decreased during 2020. Many news outlets have reported that the pandemic has given consumers more reasons to eat local food, accelerated the trend toward online grocery shopping, and changed gardening forever, reports that might partly be behind the increases in search interest in these terms as people respond to media coverage. News outlets might also write more media coverage in response to high search interest.

Although the increase in online search interest might signal increased demand for information on these topics but may or may not translate into more actual transactions by the general U.S. public, some horticultural businesses might interpret this inflow of information as a sign of steady consumer demand for related products and services beyond 2020 and decide to pivot their businesses toward these industry segments. Yet, Figures 1 and 2 combined suggest that some of these segments might have more staying power than others. While search interest in Local Food, Online Groceries, and Home Gardening has increased during 2020, Figure 2 shows that their interest peaked between March and May of 2020 and dropped considerably in the following months. Interest in Cottage Foods was relatively low in March but peaked in July. After dropping, interest in Local Food and Cottage Food has increased, interest in Online Groceries has fluctuated, and interest in Home Gardening has continued to decline. A look at Google users’ search interest since 2004 to date (Figure 1) shows upward trends for Local Food and Cottage Food, an also upward but much flatter trend for Online Groceries, and a declining trend for Home Gardening. Interest in Local Food has increased over time at a pace slower than but similar to that in Cottage Food and Food Waste (as indicated by Local Food’s flatter trend), while interest in Online Groceries has increased over time at a pace much slower than that in Local Food (as indicated by Online Groceries’ flatter trend).

If pre-pandemic trends are any indication, it is possible that search interest in Local Food, Cottage Food, and Food Waste will continue to rise after the pandemic, maybe fueled by the recent interest in short local supply channels, the expansions to some states’ cottage food laws, and the growth in the upcycled food products industry. While search interest in Online Groceries has seen an upward trend since 2004, interest after COVID-19 might not grow as fast and dramatic as 2020 levels might suggest. It is also likely that interest in Home Gardening will wane beyond 2020. In a recent presentation at this year’s virtual Southern Outlook Conference, Dr. Ben Campbell from the University of Georgia reported that the Green Industry (which includes home garden centers, nurseries, turf, floriculture, among other sectors) has grown over the past decade and during COVID-19. He finds in a survey that 60% of respondents planted a garden, put in new turfgrass, or did some outdoor renovation like putting in new plant beds because they spent more time at home during the coronavirus pandemic. But he also warned that some firms might overproduce in 2021 because they might expect 2020 demand levels. Of course, many other factors like the effects of the pandemic-induced recession on the U.S. public’s income will influence the direction of these industry segments in the future, particularly for non-essential goods.

September Seasonal Price Impacts

The past few weeks we have focused on market reactions to the Tyson beef plant fire (read again HEREHERE, and HERE). This event negatively impacted cattle prices and will remain a key topic throughout the Fall. But this is far from the only factor affecting cattle prices, especially as more time passes since the fire. One of the key factors that usually pushes prices lower during this time of year is seasonality.

We have discussed seasonal patterns in this newsletter before (see HERE). In short, they are the normal price patterns throughout a year driven by production and marketing patterns. Seasonally, either September and October are the lowest price months of the year for both fed cattle and feeder cattle due primarily to large volumes. Using the Alabama price index data above, September prices for 600-700 steers are 3% lower on average than during August and October prices are about 6% lower. The story is the same for Mississippi and most Southeastern states. It is also important to note that August is usually a little stronger than July – a bump which did not happen this year.

A common question right now is “when will prices get back to where they were before the fire?” But the story is more complicated than that given the time of year that the fire occurred: right before prices usually find a bottom. Auction prices have recovered some since the lows seen the week following the fire. Mississippi auction prices were 4 to 7 percent lower last week than the week prior to the fire depending on weight class. This is an improvement from the 6 to 10 percent drop seen in the week after the fire. Prices may not get all the way “back” partially because seasonal patterns usually negatively impact prices this time of year.

Price and Production Impacts

Special Note: News over the weekend reports that the U.S. and Japan have agreed in principle to a deal that would reduce tariffs on U.S. beef exports to Japan. There are few details now and the agreement is not formal yet, but overall it is great news for the beef industry. I’ve discussed how important a deal with Japan would be many times in this blog (see herehere, and here). More to come as more information is released…

The Tyson Fire is now two weeks behind us and we are starting to get a better picture of the impacts on markets and production. This week’s newsletter will be more visual than usual because I think these graphs tell the story well. As shown in the graph above, the boxed beef cutout value has increased sharply since the fire. This, combined with the lower cattle prices shown below, has led to some very large margins for packers which is a strong economic incentive to figure out a way to offset the loss in packing capacity.

Perhaps the first thing that most cow/calf or stocker producers notice is the impact on feeder cattle prices. The action in the futures market was clear, but so has been the drop in auction prices over the past 2 weeks.

The same declines occurred in the live cattle market shown in the graph below. The rationale here is that live cattle prices would have to be cheaper due to the temporary lower slaughter capacity.

This next graph is where things get interesting. Even with the temporary closure of the Tyson plant, total cattle slaughter last week was up nearly 2 percent from a year ago and higher than in the weeks prior to the fire according to USDA-AMS data.

There are plenty of reactions to this information. The first is likely, why did cattle prices drop so much? After all, wasn’t it the fear of less slaughter that was the culprit for the significant declines? Another reaction could be that the markets reacted in a way that kept cattle and beef moving through the supply chain.

My reaction is a little of both with the understanding that there are a lot of moving pieces. The markets certainly reacted in the directions that we should expect given a shock to the supply chain (see Dr. Jayson Lusk’s analysis on that here). The magnitude of the change is the debate – a debate that is being conducted every day in the futures markets. September feeder cattle futures closed today at $136 which is only about $2 lower than the day before the fire. This supports the idea of now that we have more information, we are not as bearish on prices.

One point to keep in mind is that U.S. packers as a whole were not running at 100% capacity pre-fire and thus there was some room to absorb some of the loss. This is also “total cattle slaughter” and not just steer and heifer slaughter so it will be worth looking at the breakouts in coming weeks. It could have been more costly processing that was not being used pre-fire such as adding extra weekend shifts, but the high margins since the fire likely made even the most costly processing “worth-it.”

All else equal, I expect beef values and cattle prices to revert to more normal ratios as time passes, especially given the evidence that the capacity is still possible.

Week in Review and a Look Ahead

Wow, what a week. The fire at the Tyson and the crop production report pushed markets into a frenzy during the first few days of the week. Fears of a slowdown in slaughter pushed beef prices higher and live cattle prices lower. This drove up the margins for packers and gave a strong incentive to absorb some of the lost slaughter caused by the fire. Choice boxed beef prices increased by about 10% percent in one week.

There was plenty of talk about overreaction of markets on Monday and Tuesday. The argument can be made that cattle prices should not have dropped as much as they did. The fact that futures prices rebounded on Wednesday supports that idea. But markets don’t have to be “exactly right” in the short-term to still be efficient in the long-run. Market adjustments are typically very subtle but when a shock such as this fire occurs, wild swings happen.

Markets react to news in a way that allows products to go to where they are most demanded. In this case, the higher packer margins incentivized plants to increase slaughter in order to offset some of the loss from the Kansas plant. This is a primary reason that weekly slaughter totals will not be 5,000 head lower while the plant is closed even though that is the report of what the Tyson plant was processing each week.

I do think this is a short-term shock, but a significant shock. The underlying fundamentals of the beef sector have not changed so much that we should expect this impact to persist in the long-term, especially once the plant comes back on board.

 

I want to close with the following is an excerpt from Dr. Derrell Peel at Oklahoma State which I think is very well-written.

“The complex set of markets in the cattle and beef industry are all impacted initially.  Live and feeder futures dropped sharply for two days before beginning to moderate late last week.  It is one of the functions of the futures markets to anticipate the worst case scenario, especially in the face of much uncertainty, before moderating as the reality of the situation becomes clearer.  Feeder cattle markets also decreased in the face of lower fed prices and the uncertainty roiling all markets.

Is the initial reaction an overreaction?  In one sense yes, but it is a very common market response to reestablish supply and demand balance quickly.  We see it in all markets and certainly in agricultural markets. Corn prices of $7/bu. in 2012 helped ensure increased corn production to overcome drought impacts and meet growing corn demand; and $3/cwt. calf prices provided the temporary incentive to jumpstart herd expansion in 2014.  By virtue of extreme initial reactions, markets ensure that equilibrium in supply and demand is reestablished as quickly as possible. The sudden shock of the current situation and the resulting big initial market reactions encourage buyers and sellers to change plans; incur additional costs; and react quickly to new arbitrage opportunities.

What should we expect going forward?  Similar situations from the past may provide some indications.  In December, 2000, a ConAgra beef packing plant in Garden City burned completely; never to reopen.  Subsequent research confirmed initial reactions generally similar to the current situation.  Most of the negative impacts on fed cattle prices subsided in three to six weeks after the event.  Packing capacity relative to cattle supplies is somewhat tighter this time so the impacts may be slightly larger or longer-lived.  Nevertheless, boxed beef and cattle markets will likely adjust relatively quickly in the coming weeks with final adjustments depending on the duration of the plant closure.”

Some Hard Data on Lab-Grown Meat

Lab-grown meat (and plant-based meat) has been a major topic of discussion in cattle groups over the past few years. Most of this discussion has centered around questions and speculative answers. The first questions centered around “Can beef really be grown in a lab?” The answer to that one has proven to be yes. The next logical question is “how much will it cost?” I’ve seen plenty of guesses but nothing concrete. The most interesting and perhaps most important question has always been “will anyone actually eat it?” Newly released research by Dr. Jayson Lusk at Purdue University and others provide some insight into this question as detailed in his blog post last week.

“With all the news about Beyond Meat’s stock price and the rolling out of the Impossible Burger at Burger King, there has been a lot of speculation about how consumers might response and about the ultimate size of this market. In a new paper with Ellen Van Loo and Vincenzina Caputo, I’m pleased to bring some hard data to the these debates.

What did we do? We surveyed about 1,800 U.S. food consumers earlier this year and asked them to make a number of simulated shopping choices. In each choice, consumers had five options: conventional farm-raised beef, a plant-based burger made with pea protein (i.e., Beyond Meat), a plant-based burger made with animal-like protein (i.e., Impossible Foods), labgrown meat (i.e., Memphis meats), or they could choose not to buy any of the products (i.e., “none”). Respondents were randomly allocated to different treatments that varied the use of brand names (present/absent) and the information that was provided (none, environment information, or technology information).”

What did they find? At constant prices and conditional on choosing a food product, “72% of respondents chose farm-raised beef, 16% plant-based (pea protein) meat alternative, 7% plant-based (animal-like protein) meat alternative, and 5% lab-grown meat. Adding brand names (Certified Angus Beef, Beyond Meat, Impossible Foods, and Memphis Meats) actually increased the share choosing farm-raised beef to 80%.”

“Even if plant- and lab-based alternatives experienced significant (e.g., 50%) price reductions, farm-raised beef maintains majority market share. Vegetarians, males, and younger, more highly educated individuals tend to have relatively stronger preferences for the plant- and lab-based alternatives relative to farm-raised beef. Respondents are strongly opposed to taxing conventional beef and to allowing the plant- and lab-based alternatives to use the label “beef.”

Lusk closes by saying “Because these are new products just hitting the market, it is possible that these preferences can and will change, particularly when more consumers are able to taste them. However, at present, the future market potential for these products appears to fit more in the “niche” category, even at significant price discounts. What will happen in the future? Only time will tell.”

 

Cattle Cycle Dynamics

There were no big surprises in the recent July 1 Cattle Inventory report. The beef cow herd was estimated to be the same as a year ago while the number of heifers held back for replacements was lower than a year ago. These estimates were in line with what was expected following the January report.

There has been plenty written about the current cattle cycle topping out and this report was another piece of evidence supporting that story. The July report marked the fourth consecutive January or July report that showed less than one percent growth compared to the previous year. While still not a decline, these modest or flat growth numbers are a contrast to the rapid growth from 2015-2017.

This is a natural time to look at the longer-view state of the cattle and beef cycle. The July story from 2017 to 2018 was larger calf crop but fewer heifers retained. The story from 2018-2019 is a small decline in the calf crop and even fewer heifers retained. The estimated 2019 calf crop of 36.3 million head would be the first calf crop decline (albeit a small decline) when compared to the previous year since 2014.

Cattle on feed are at record large levels that would be expected at the peak of a cycle. However, dressed weights have moderated over the past few years and especially in the past few months which has softened some of the impact of larger cattle inventories on total beef production. June marked the ninth consecutive month that federally inspected steer dressed weights were lower than the previous 5-year monthly average. This is pretty rare given that dressed weights have been increasing for decades but not too unusual when in the context of the current cycle. The same streak of declines occurred in 1996-1997 which was the peak of the 1990-2004 cattle cycle and there was also a similar period in 1984 which was near the peak of the 1979-1990 cycle.

While it is fairly clear that the cattle numbers have leveled out (at least for now), the story is different for beef. The natural lags in the cattle and beef sector cause beef production to peak later than cattle do. Beef production will continue to be bigger for the next few years while these supplies work through the beef supply chain.

June Feeding Closeouts Dip into the Red

Cattle feeding closeouts turned negative in June for the first month in 2019. The Livestock Marketing Information Center (LMIC) calculations show the June closeout was -$60.66 per animal sold for feeding-out a 750-pound steer in a commercial Southern Plains feedlot. Similarly, calculations by Iowa State University (ISU) for steers placed as yearlings and calves and sold at slaughter-weight in June were also negative. ISU calculated that steers placed as yearlings and sold in June generated a loss of -$39.71 per head (including manure credit), and animals placed on-feed as calves came in at -$100.61. For yearlings, the LMIC calculations showed that before June, the last red ink was for September 2018, while ISU last estimated a negative profit in December 2018.

Neither the LMIC nor ISU estimates are survey-based. But they do provide indications of the direction of change. Of course, late 2018 and the first several months of 2019 many cattle feeders had much worse results than these calculations, which are based on normal weather. Very muddy feedlot conditions resulted in red ink for many cattle feeders. The baseline production systems and assumptions for the LMIC and ISU are different. Besides using different prices and costs, ISU incorporates, for example, Modified Distillers Grains in the ration, which is a common feedstuff there.

The LMIC makes projections for breakeven sales prices in coming months based on recent feedstuff costs, etc. A steer reaching market weight in July (placed at 750-pounds) has an estimated breakeven sales prices of about $111.25 per cwt. August’s breakeven level is projected to be similar to July’s. So, closeouts for July and August are likely to bring more red ink. If feedstuff costs do not skyrocket, cattle feeders are expected to generally breakeven or post small profits late this year. In the situation where corn cost is already locked-in, November breakeven sales price is in the range of $105.50 to 106.50 per cwt, and December is $111.00 to $112. Current Live Cattle futures prices (shown below) are about $109 for the August contract and $114 for December.

This article was written by the Livestock Marketing Information Center with some additional information included. LMIC is a cooperative effort between state university extension specialists (including me), USDA economists, industry cooperators, and LMIC staff. For more info about LMIC, click here.

Level U.S. Cattle Inventory

USDA-NASS released the mid-year cattle inventory report last Friday. This is an annual report that estimates the number of cattle on July 1 each year. It is not as comprehensive as the January 1 report in that there is no state-by-state breakout. However, the information in the report is a useful look over time to get an estimate of the cattle inventory cycle.

The all cattle and calf count was unchanged from a year ago (103 million), as was the number of beef cows (32.4 million). This estimate supports the idea that the rapid cattle expansion seen in the past 5 years has leveled-off.  The July 1 inventory of steers over 500 pounds was 14.7 million head, up 1.4 percent year over 2018 levels.  The inventory of other heifers over 500 pounds was 7.9 million head which is up about 5 percent.  Overall, this report was well-anticipated and contained no surprises.

Perhaps the most interesting estimates were the number of heifers reported as being held for beef cow replacement purposes and the 2019 estimated calf crop. The number of heifers reported as being held for beef cow replacements declined by just over 4 percent. This is another sign that herd expansion has slowed. The first estimate by NASS for the 2019 calf crop showed a slight drop compared to 2018 at 36.3 million head compared to 36.4 million head a year ago.

What does all of this mean? My take is that the cattle inventory numbers are in in a bit of a holding pattern, there is not a lot of evidence (i.e. really low prices) to suggest that heavy liquidation is soon to come. There is also not a lot of motivation (i.e. high prices) for continued expansion. I expect the rest of 2019 and into 2020 to continue to be mostly flat on inventory changes.