Five weeks into the full-blown corn demand destruction in the ethanol industry as a result of the coronavirus pandemic’s impact on gasoline demand, the market will be heavily focused on the USDA’s reassessment of the situation. Whether or not USDA continues to hold onto soybean export optimism will be watched closely, as well. The market is preparing for what could be/likely to be a new crop corn ending stocks projection unlike anything seen in decades as the USDA issues their first official 2020/21 balance sheet projections, as well.
A summary of the average trade estimates is attached as a PDF file at the end of this post.
Corn
The hot topic of interest regarding the old crop corn balance sheet will obviously be the ethanol situation. While we have seen some stabilization in U.S. ethanol production in recent weeks, current run rates remain substantially below USDA U.S. corn demand assumptions. Over the last four weeks, U.S. ethanol production has run, on average, 45.0% below year ago levels, resulting in an estimated 186 million bushel loss of corn demand relative to last year during the same period. We estimate 2019/20 marketing year to date corn used for ethanol production is down roughly 225 million bushels from last year at this time, while the USDA’s current 5.050 billion bushel annual estimate reflects an expected 328 million bushel total decline from last year’s 5.378 billion. If ethanol production continues to run 40% below year ago levels, the USDA’s estimated annual decline in corn for ethanol usage will be reached in roughly two weeks. While there have been some signs of life in rebounding gasoline demand, it is nowhere near levels implying an end is anywhere in sight for continued considerable reductions in ethanol production.
To put some further perspective on the situation, in order for 2019/20 corn for ethanol usage to achieve the USDA’s current 5.050 billion bushel estimate, ethanol production would need to run, on average, roughly 6% below year ago levels through the endo of August vs the 45% declines experienced in recent weeks. Cutting 100 million bushels from the USDA’s projection would require ethanol production running roughly 11.5% below last year from this point forward, 200 million bushel cut (4.850 billion) a roughly 17% reduced ethanol production rate and 300 million bushel cut (4.750 billion) a roughly 23% ethanol production decline through August. As seen, the downside risk in 2019/20 corn for ethanol usage is still rather substantial, even following last month’s much larger-than-expected 375 million bushel cut. Given the bludgeoning of ethanol production over the last month, many would be ecstatic with ideas/prospects for ethanol production to rebound to just 15% year-over-year declines over the last four months of the marketing year, but even that assumption could add another 175 million bushels to old crop ending stocks. We’re currently estimating 2019/20 corn for ethanol usage at 4.825 billion bushels and leaning lower with every week going forward in which ethanol production runs 30%+ below year ago levels.
On the export front, while the market’s attention seems to continuously be on constant newswire reports of Asian countries buying South American corn, export demand for U.S. corn over the last two months has been extremely impressive. In fact, over the last eight weeks, U.S. corn export sales averaged 49.7 million bushels/week, which is an all-time record for the period. Last year’s corn sales during the same time frame were just 27.2 million bushels/week. Given the strong pace of sales of late, corn sales during May-August would only need to average roughly 12.3 million bushels/week in order for 2019/20 exports to reach the USDA’s 1.725 billion bushel projection. While last year’s sales during the period averaged only 8.5 million bushels/week, May-Aug corn sales over the last five years averaged 16.4 million bushels/week, with the previous two years of historically strong March-April sales (2017/18 and 2015/16), as was the case this year, seeing May-Aug sales average 21.1 million and 26.8 million bushels/week, respectively. Any glimpses of Chinese demand, as has been rumored of late, would only further bolster 2019/20 export ideas. Clearly the 2019/20 export program is going to go down as being disappointing in historical terms, but we feel the USDA’s projection now appears a bit on the side, although nowhere near levels to offset the massive demand losses of the ethanol situation. We’re estimating 2019/20 U.S. corn exports at 1.775 billion bushels, 50 million above the USDA.
The old crop feed/residual situation is a fairly complex one with multiple considerations amid the coronavirus pandemic prompting various thought processes. On one hand, the widespread meat processing plant closures/slowdowns are forcing are higher than usual amount of animals to be kept on feed, potentially raising feed usage ideas. On the other hand, as animals move beyond ideal weights given the reduced slaughter capacity, feed rations are being reduced to slow weight gain, as well as to reduce costs. Additionally, animal number reductions are increasingly being seen, either through natural ways (reduce placements/increase marketings) or by forced methods without other options. April cattle on feed was down 5.5% from last year, with March placements down 23% from a year ago. Egg sets and chick placements have been down 6-8% in recent weeks, while reports of euthanizing hogs have been widespread. Anecdotally, reduced feed demand of late is implied by recent comments by Bunge’s CEO of U.S. soybean crush being reduced by around 10% due to such influences. Quite frankly, we’re unsure of how USDA may address this situation in next week’s WASDE report, as they have the option of simply waiting until the June 30 Grain Stocks report to get the next read on quarterly feed/residual usage. We have trimmed our 2019/20 feed/residual usage estimate to 5.600 billion bushels, 75 million below the USDA’s 5.675 billion bushel estimate at this time.
The one other unknown for the 2019/20 U.S. corn balance sheet is the USDA’s resurveying of northern corn belt states regarding last year’s crop following the inability to complete harvest last fall. USDA’s first round of resurveying will be reflected, if needed, in the May 12 Crop Production report. For this round, USDA resurveyed producers in MI, MN, SD and WI for any updated harvested acreage and production numbers relative to those provided for the January Annual Crop Production report. As of the USDA’s final harvest progress update last fall as of December 8, MI corn was 74% harvested, leaving 62 million bushels unharvested, MN 93% harvested (88 mil bu unharvested), SD 83% harvested (96 mil bu unharvested) and WI 74% harvested (23 mil bu unharvested) for a total of 269 million bushels still in the field at the time. Those bushels also counted towards on-farm stocks for the USDA’s December 1 Grain Stocks estimate of 11.402 billion bushels. Unfortunately, none of these four states have provided corn harvest updates this spring in their weekly crop progress reports of the states’ activities for the week unlike North Dakota has. Using the North Dakota situation as a proxy, even as its situation was much worse than these four states, we don’t feel there is likely to be a large-scale reduction in last year’s crop estimate and/or Dec 1 stocks reflected in next week’s report as the ND harvest has advanced to 91% complete as of last Sunday vs just 43% complete as of Dec 8. Obviously there is the unknown of what portion of the crop may have been lost/damaged since last fall, along with quality concerns (which do not play into an outright bushel count per se), but we don’t expect it to be significant, particularly in the context of the 500 million+ bushels in lost corn demand in the balance sheet from the ethanol situation. USDA will resurvey North Dakota farmers specifically at another time given the magnitude of the situation last fall. As mentioned, as of Dec 8, just 43% of the ND corn crop was harvested, leaving 260 million bushels in the field at the time. Based on their latest update of harvest being 91% complete, around 41 million bushels remained unharvested as of last Sunday. Hopefully, USDA will provide an update in next week’s report as to when the ND resurvey will take place.
Putting it all together, we’re estimating 2019/20 U.S. corn total demand at 13.610 billion bushels vs USDA’s last estimate of 13.865 billion and last year’s 14.288 billion, resulting in our ending stocks estimate of 2.347 billion bushels vs USDA last at 2.092 billion and last year’s 2.221 billion. Our ending stocks estimate puts the 2019/20 stocks/usage ratio at 17.2%, the highest since 2005/06, with outright ending stocks the highest since 1987/88. USDA’s last ending stocks estimate was 2.092 billion bushels with a stocks/usage ratio of 15.1%. The average trade estimate of 2019/20 ending stocks is 2.261 billion bushels.
USDA will be issuing their first official estimate of the 2020/21 U.S. and world balance sheets in this month’s WASDE report, as well. While they have provided some preliminary views of the new crop situation at the February Ag Outlook Forum and in the 10-year baseline projections, these are the first official estimates. For all intents and purposes, how the USDA will reflect the 2020/21 supply side of the U.S. corn balance sheet is known as they will assume the March Prospective Plantings report estimate of 96.990 million acres planted and the February Ag Outlook Forum yield assumption of 178.5 bushels/acre. This has been their standard procedure for years, never deviating from March acreage estimate (even last year) in the May WASDE, and assuming the February yield number in each of the last six years. With planting off to a very quick start, we do not anticipate USDA doing anything out of the ordinary in this regard. Accordingly, their 2020/21 U.S. corn crop estimate should come in around 15.945 billion bushels (minor unknown on harvesting percentage assumption) vs last year’s 13.692 billion – a 2.25 billion bushel increase. At the time of the February Outlook Forum, USDA was estimating old crop ending stocks at 1.892 billion bushels, but are now expected to be estimated at 2.2-2.3 billion bushels by USDA, which would put 2020/21 total supplies at more than 18.2 billion bushels vs 2019/20’s 16.0 billion, 800 million bushels higher than USDA reflected in February, when they were already estimating 2020/21 ending stocks at 2.637 billion bushels. Accordingly, we see the potential for USDA to reflect new crop ending stocks above 3.4 billion bushels. We’re currently working with a 2020/21 U.S. corn ending stocks estimate of 3.261 billion bushels, based on planted acreage of 95.5 million acres and old crop ending stocks of 2.347 billion. The average trade estimate of 2020/21 U.S. corn ending stocks is 3.420 billion bushels.
Soybeans
Starting with the export situation for soybeans, unlike corn, despite a few weeks here and there of decent sales numbers, the export program overall, of late, has been far from impressive, particularly in context of the USDA’s annual export ideas. Over the last eight weeks, even with occasional bouts of Chinese buying, U.S. soybean export sales averaged 22.9 million bushels/week, below last year’s 27.0 million/week average during the same period, as well as 2017/18’s 31.1 million/week. Given the very slow start to the 2019/20, the attempt to play “catch up†is proving to be quite difficult as total commitments of 1.435 billion bushels still only reflect 80.8% of the USDA’s estimated annual exports, while at no time have late April sales on the books reflected less than 87% of eventual annual exports, leaving the USDA’s 1.775 billion bushel projection appearing quite optimistic.
In fact, if the USDA’s current export projection is to be met, May-August soybean sales would need to average roughly 19 million bushels/week through the end of August. For comparison, last year’s sales from this point forward averaged 7.6 million bushels/week and were 8.2 million and 7.8 million/week in the previous two years. Record May-August average weekly soybean sales are 13.6 million bushels/week, set in 2015/16. In order to get the average “needed†sales pace from this point forward to a level consistent with past years’ typical purchases by non-Chinese destinations, we estimate China would need to buy at least another 6-7 MMT of old crop soybeans if the USDA’s 1.775 billion bushel projection has any chance of being met. China’s timing and motivations in buying U.S. soybeans are highly unknown at the moment. By all means, it’s possible China will step up and take advantage of the currently-depressed prices to rebuild state reserves, with the Phase One trade deal agreement obviously paving the way for large-scale purchases “needing†to be made, but the data we currently have suggests the USDA’s export projection should be lowered. We’re currently estimating 2019/20 soybean exports at 1.725 billion bushels, 50 million below the USDA.
The soybean crush situation is a tale of two stories. Known data so far implies the USDA’s 2.125 billion bushel annual estimate is too low. March crush of 192.2 million bushels obliterated the previous record for the month of 182.2 million bushels, with each of the last four months’ crush rates setting a new record for the respective month. Over the last three months of official data (Jan-Mar), crush ran 6.0% above year ago levels, with 556 million bushels being crush, 31 million bushels more than during the same period last year. The USDA’s annual crush estimate reflects an expected 33 million bushels increase for the year.
However, crush margins through the end of March consistently ran at or above year ago levels. Since the start of April, though, crush margins weakened, with July 2020 board crush (red line in chart below) now at 87 cents/bushel vs $1.21/bushel on the same date last year (blue line) and $1.70/bushel in 2018 (black line). Additionally, there have been industry reports of crush rates being reduced amid declining feed demand due to impacts of the coronavirus pandemic. In order for the USDA’s 2.125 billion bushel crush estimate to be reached, April-August crush would need to total 860 million bushels, nearly 1% above last year’s 852 million. As seen in the chart above, last year’s April, July and August crush rates were all records for the month, though May and June soybean crush were well off record levels. Moreover, 2018/19 April-August crush holds the current record for the period at 862 million bushels, implying crush over the final five months of the marketing year needs to match the record in order to reach the USDA’s target. For these reasons, we have pulled back our 2019/20 crush estimate to 2.115 billion bushels, 10 million below the USDA.
July Board Soybean Crush (2020 RED, 2019 BLUE, 2018 BLACK)
USDA also resurveyed soybean producers in MI, WI, MN and SD for this month’s report and will provide updated 2019 production numbers and December 1 stocks numbers if the survey results warrant revisions. At the time of the final soybean harvest update last fall, as of December 1, the MI crop was 85% harvested (11 million bushels unharvested), WI 86% harvested (11 mil bu unharvested), MN 99% harvested (3 mil bu unharvested), while SD was 100% harvested. Accordingly, the total “at risk†production from these four states is roughly 25 million bushels. The eventual resurvey of North Dakota farmers will provide an update, if necessary, on the 8% of unharvested area as of December 1, or roughly 14 million bushels in production. We do not anticipate the resurvey reflecting a considerably altered view of last year’s production as a maximum of 40 million bushels appears to be “in play,†with the vast majority of those bushels likely having been harvested at some point.
Based on our lower export and crush estimates, we see 2019/20 U.S. soybean ending stocks at 524 million bushels vs USDA last at 480 million. The average trade estimate is 497 million bushels.
For the USDA’s first official look at the 2020/21 balance sheet, as with corn, we fully anticipate USDA assuming the March Prospective Plantings acreage estimate of 83.5 million acres and the February Ag Outlook Forum average yield assumption of 49.8 million bushels. Accordingly, we anticipate USDA’s 2020/21 crop estimate to be 4.121 billion bushels vs the 4.195 billion they reflected at the February forum as they were looking for soybean area at 85.0 million acres at the time. However, USDA was using an old crop ending stocks estimate of 425 million bushels at the time vs 480 million last month by USDA and our views of 524 million. Accordingly, assuming 2020/21 beginning stocks around 500 million bushels, the higher “starting point†offsets the expected lower crop estimate to keep total supplies nearly identical to the USDA’s assumption in February of 4.635 billion bushels. With the similar total supply outlook, USDA’s 2020/21 ending stocks estimate in February of 321 million bushels may not see much change in their first official estimate. Our current view of the new crop situation is for modestly higher acreage at 85.0 million, along with slightly better demand expectations than USDA, but also with higher beginning stocks to put our 2020/21 U.S. soybean ending stocks estimate at 386 million bushels. Taking the March acreage estimate verbatim, though, would put our ending stocks estimate at 312 million bushels. The average trade estimate of 2020/21 soybean ending stocks is 440 million bushels.
Wheat
With the 2019/20 U.S. wheat marketing year entering its final month, we’re not anticipating any notable changes by USDA at this time. There continues to be some debate over final export numbers, as the export sales pace continues to run a levels more than adequate to all the USDA’s 985 million bushel projection to be met, but the actual shipment pace has been far off “needed†levels. In recent years, it has been common for wheat loadings/shipments to increase notably in the final weeks of the marketing year as old crop commitments are fulfilled so we’re not writing off the likelihood of the USDA’s estimate being reached, but a fairly immediate pick up in exports will be needed.
Based on the level of export sales already on the books, the USDA’s 985 million bushel projection appears likely to be met. However, based on cumulative Export Inspections through April 30, and taking into account the difference between official Census Bureau export data and Inspections data, while also accounting for flour/wheat product exports, we estimate wheat export inspections over the last four full weeks of 2019/20 will need to average roughly 25.7 million bushels/week vs the 17.3 million/week average seen over the last eight weeks. Similarly, last year’s March-April average weekly wheat exports were 19.9 million bushels/week and finished the final four weeks of the year averaging 26.1 million/week. A repeat of that will be needed in order to reach the USDA’s 985 million bushel export projection. We’re using 980 million in our balance sheet.
We’re currently estimating 2019/20 wheat feed/residual usage at 140 million bushels vs USDA last at 135 million, with our old crop ending stocks estimate of 973 million bushels in line with USDA’s 970 million, with the average trade estimate at 969 million bushels. Unless something unexpected is reflected, the focus for wheat this month will be heavily on the USDA’s first objective estimate of the 2020/21 U.S. winter wheat crop.
Over the last several weeks, U.S. winter wheat crop conditions have declined moderately, taking top-end yield expectations off the books after early April conditions were nearly identical to those of a year ago and at the top end of conditions over the last 10 years. Heading into next week’s Crop Production report, overall winter wheat conditions, while still above the most-recent 5-year and 10-year averages, are solidly below those of recent top-yielding years of 2019 and 2016.
In addition to the yield uncertainty for the May estimate, there is also considerable uncertainty regarding harvested acreage as this month’s data will provide the first official estimate. While harvesting percentage is typically impacted by the condition of the crop, with this year’s “average†conditions not implying anything out of the ordinary should be seen, over the last 10 years, the winter wheat harvesting percentage has ranged from 75.5% to 85.4%. Based on the USDA’s 30.8 million acre planted area estimate, harvested area anywhere in the 23.2-26.3 million acre range technically is “in play†and reflecting as much as a 160 million bushel range in production between the two harvested acreage scenarios based on current yield ideas. We’re assuming a fairly typical harvesting percentage of 78.5% in our crop assessment heading into next week’s report and compares to last year’s 78.1% and the most-recent 5-year average of 79.3%. As a reminder, USDA estimated U.S. winter wheat planted area at 30.775 million acres vs last year’s 31.159 million. Our harvesting percentage assumption puts harvested area at 24.170 million acres vs 24.327 million last year.
We’re estimating the 2020/21 U.S. all winter wheat crop at 1.259 billion bushels vs 1.304 billion last year, with an average yield of 52.1 bushels/acre vs 53.6 last year. We see the hard red winter wheat crop at 730 million bushels vs 833 million last year, soft red winter wheat at 303 million vs 239 million last year and winter white wheat at 226 million vs 232 million last year. The expected decline in HRW production from last year is due to a combination of lower yields and lower acreage (see summary acreage table below), while the increase in SRW production is from just the opposite, with solidly higher yield expectations on much better crop conditions than a year ago, along with an increase in acreage reflected by USDA in the Prospective Plantings report. The average trade estimates are: all winter 1.245 billion, HRW 739 million, SRW 280 million, white 226 million.
Using the USDA’s spring and durum wheat planted acreage estimates, along with average yield assumptions, we’re working with a 2020/21 U.S. all wheat crop estimate of 1.873 billion bushels vs last year’s 1.920 billion, but above the USDA’s February Ag Outlook Forum assumption of 1.836 billion. Our current other spring/durum wheat production assumption of 614 million bushels compares to last year’s 616 million bushels. USDA does not provide their first objective estimate of the spring wheat and durum crops until the July report.
Based on our HRW crop estimate, we see 2020/21 U.S. total supplies at 1.246 billion bushels vs last year’s 1.354 billion, while SRW total supplies are estimated at 421 million bushels vs 402 million last year.
We would note there has been absolutely no bias historically in the USDA’s May winter wheat crop estimate relative to market expectations, with four of the last seven years’ reports coming in within 10 million bushels of the average trade estimate, but there have been plenty of surprises of 50+ million bushels over the years, as well.
Looking at the 2020/21 all wheat balance sheet, we’re currently looking for another modest decline in ending stocks to 832 million bushels from this year’s 973 million, reflecting the 4th consecutive yearly decline, but still not at levels that would be considered fundamentally bullish. Moreover, our new crop ending stocks ideas are moderately above the USDA’s February Ag Outlook Forum assumption of 777 million bushels. We’re looking for moderately better U.S. wheat exports over the coming year, but reduced domestic/feed usage as a result of the anticipated massive corn crop. The average trade estimate of 2020/21 U.S. wheat stocks is 818 million bushels.
