Starting with the weather; I’m not sure the last time that I’ve seen such a friendly drought monitor in mid-July. From a midwestern perspective, it’s not just affecting row crops, either. I was home in Southwest Missouri for the 4th, and there was a lot of baling going on. I don’t ever recall seeing that density of bales in the fields in July. If the past decade’s cattle herd numbers have been held back by some combination of balance sheet, profitability, or pasture and forage availability, that won’t be the case in 2025. I expect that the beef herd is already growing, and this will hopefully provide some additional grain demand in the coming years. In the July WASDE, feed use (across all species) is forecast to rise about 120m bushels when combining corn, wheat, sorghum, barley and oats. I believe that this is too low; and expect to see another 25-40m bushels of feed demand added as we go through the marketing year.

Wheat exports remain a surprise this year. While market conditions (lower prices, weaker USD, and lower EU exports) and USDA reports have continued to project higher exports, the market continues to run ahead of even these projections. The July report projects 850m bushels of exports in the 25/26 marketing year, but historically, we are on pace for 950m bushels. It is still relatively early in the year to go wild with upwards revisions, but I expect to see the August exports to increase by 25m bushels if we see this pace continue.

Soybean export pace is worrying. Between the large South American crop, and the uncertainty surrounding contracted pricing six months hence, the weakness in prices and the USD haven’t been able to result in higher exports. In better news for the soy complex, the OBBBA does place significant new restrictions on 45z tax credit eligibility. Starting in 2026, only feedstocks from USMCA countries will be eligible for 45z credits. This final language resulted in a huge boost in soybean oil prices as domestic tallow and UCO supplies are simply not large enough to meet the already existing demand.

Soybean export sales for the 2025-26 crop year remain very weak compared to historical averages. This isn’t a surprise. I still believe that the market harbors lots of concern about the trade tensions between the US and China. Further, Brazil’s continued availability from this year’s Brazilian, and global, record production has dampened demand. I think it’s far too early to get extremely concerned—as one can see from the plot above, there have been years in which a start this slow did recover to meet USDA export estimates. However just because we are all tired of talking and hearing about trade tension doesn’t mean that tension isn’t real.

Read more

I’ve seen a number of charts similar to this on social media in the past few weeks, implying that corn prices typically break down in mid-June. While I certainly don’t disagree that its common, I do think its a good time to remember the words of the famous 20th century physicist Murray Gell-Mann, “Physics would be really hard if particles could think.”

Price patterns are not physics. There is no gravity holding them together or atomic forces pushing them apart. The ‘Laws’ of supply and demand are neither enforced nor enforceable.

What we see above are repeated patterns. Humans are incredibly good pattern recognizing machines. We are so good that we often recognize patterns that aren't there. That’s why we need more than patterns and technical type indicators. We need to understand what is going on to be able to deduce the cause of the patterns, and make guesses about how often they’ll recur.

If you mouse over the chart above, you’ll see how many years of December futures data are used make each average. The 5 year line is the average of 2020-2024 years. I’ve arbitrarily normalized the graph so that each June 15th, prices are 4.39, roughly the current price of corn.

Let’s start with this, why would these patterns exist in new crop corn? Simply because June and July are when a lot of uncertainty about the condition of the crop is created and resolved. Whether it is very fast planting pace in May or heavy rains that push planting into June, both affect prices. In July, will it be extremely dry in critical growing areas, or that perfect mix of hot days, cool nights, and twice weekly rains? As this critical 60 day period (June-July) passes, the eventual yield comes into focus. Most years, yield is at or above trend line projections, and as the worry of a short crop declines, so do prices. But some years, things go from good to bad, or bad to worse, or even good to bad to even worse.

As Don Close would ask “So What?” Scroll over the plot above, and some things stand out. Periods with the largest price declines after June 15th are also those with the largest run-up in prices leading to June 15th, which makes sense. Also, the longer the period examined, the less the change leading up to or away from June 15th.

Therefore, don’t sweat it too much. Yes, we are near the end of the crop year. Storing corn (or soybeans or wheat) across crop years is nearly always a losing proposition, so those with old crop bushels should be getting them priced sooner rather than later. If you are going to bet on a big drought in the US, the best way to do that is to purchase out of the money call options, not hold on to physical corn.

New crop corn sales are looking solid so far this year, as long as you understand the difference between the average and the median. Each week, I calculate export commitments as a percent of the latest WASDE export estimate. This allows us to examine the to-date pace of exports in the graph above. The dotted lines represent the fastest and slowest pace over the past 20 years while the darker areas are the 25th and 75th percentile.

Based on the current percent committed and the historical pace, we can calculate to what extent we are ahead or behind on pace. Looking at new crop corn above, we have contracted 4.7% of total USDA estimated exports this week, compared to an average of 7.5%, which puts us pretty far behind. With data, it’s critical to remember that one single statistic rarely tells the whole story, which is why I include the 25th and 75th percentiles. Even though the pace of sales is behind the average by a huge amount for this time of year, it’s still faster than most years, i.e. it is faster than the median. Plus, it’s still really early. Right now, in other words, I’m much happier with corn than soybeans.

Read more

Total booked sales for old crop remain ahead of the historical pace needed to hit the 2.6 billion bushels forecast in the May WASDE—in fact only 36.4 million more bushels of sales are needed to hit the WASDE number over the remaining 11 weeks. In fact, the pace that sales are being booked implies that export estimates should be closer to 2.75 billion bushels. This all depends on whether those bushels will actually get shipped.

We know that the Brazilian safrihna crop is going to be large and harvest starts this month. While much less common in corn, in soy, big harvests can lead to changes in origin of supply contracts, and therefore cancellation of US exports. The first canary in the coal mine has started coughing, and you can see why by comparing the pace of commitments (above) to outstanding shipments (below).

Read more

If you are wondering how new crop exports are looking so far with all of the ups and downs in tariffs, here is your update through the week of 22 May 2025. Looking at the graphs above, the dotted lines are max/min pace for last 20 years. The dashed lines are 25th percentile & 75th percentile pace. The red line is pace as of 22 May, the most recent data available. The US has typically contracted 8.6% of our eventual total soybean volume at this point in the year, we are currently at 2.1%. Only about 5 of the past 20 years have been slower. Corn we are 4.4% instead of 7.3%, while we are behind the average pace, we are ahead of the median pace--about 2/3 of the last 20 years were slower than this year.

What's the takeaway? New crop beans are almost certainly seeing some effects, but it's much too early to know how it's all going to play out. Meanwhile, corn is actually being booked at a pretty normal pace.

The charts above are static images, the charts below are updated each week and you can mouse-over to see the current values:

Read more

In the wake of the largest tariff increase in the US since, well, a really really really long time, I’m seeing silly posts like this one above pop up more and more.1 The implication is that because this manufacturer uses domestic steel & aluminum in their products, they are immune to price increases from tariffs. But this is simply wrong.

Ostensibly, the purpose of the tariffs is to bring back manufacturing to the US. This means that pre-tariff, foreign manufacturers could produce at lower costs than the US, hence too little being made in the US.

So, the US imposes tariffs on foreign producers, and manufacturing goes up in the US, right? But what is that mechanism? It’s price. Tariffs increase the price of the imported good enough that domestic producers can raise their prices to profitable levels, incentivizing the domestic producers to build more factories, etc.

Let’s put this another way. Say you make hammers. It costs $4 to make a hammer in Mexico and $5 to make a hammer in the US, so hammers are all made in Mexico and imported here, and sold at $4+markup. In order to spur the domestic hammer industry, the US imposes a 50% tariff on imported hammers. It now costs $6 to make and import a hammer from Mexico but only $5 for a US hammer. We stop importing hammers and make them here for $5+markup. What’s the net result? Even though the tariff only was applied to Mexican-made hammers, domestically produced hammers became more expensive. This is why economists view tariffs as just another form of a tax.

One additional advantage of this method is that it provides a natural excuse to write down the questions as they come in, which is a great record to have afterwards. Unfortunately, I didn’t write down the questions that came at the end of the presentations, which I wish I had.

31 January 2025, Community Bankers Association of Ohio, Findlay, OH. Topic was outlook for the Agricultural Economy.

• Should the US be more involved in helping developing countries to counter China’s Belt & Road initiative?

• What is the biggest thing that Ag Lenders should keep on their radar?

• As someone who recently worked in the Farm Credit system, and has now returned to community banking, what advantages do I believe that community banks have compared to Farm Credit Associations?

• What should crop farmers diversify into? What other varieties (IP, etc) are attractive, or even other businesses for young farmers?

• Can you expand on your comments on Farm Family Mental Health, Divorce and Dynamics? (This came after the presentation in which I mentioned these issues)

• What is happening to the value of the US dollar, why, and what is the outlook? Will it affect exports?

• What are my thoughts on farm land prices?

• What can or should be done to make succession planning more successful?

3 February 2025, Iowa Soybean Association Farm Forward Event: (note that this was the Monday morning after Trump had announced 25% tariffs on Canada and Mexico)

• Should we worry about tariffs? Are they just a tool in negotiations?

• Under the USMCA are the proposed tariffs legal?

• If Mexico chooses to escalate & retaliate, what effect would that have on corn prices in the US?

• Since the 1980s, there have been many policy shifts and shocks…are the tariffs any different?

• What lessons can we learn from 2024?

(EDIT: The below questions also came from the Iowa Soybean event via text. I forgot to include them in the original list)

• Did the Trump 1.0 tariffs generate more revenue for than the treasury than the cost of the direct payments given to farmers?

• With all the money China has spent in South America on infrastructure. how much effect will a tariff actually affect United States growers

• Do direct payments actually hurt smaller farmers in the long run?

• I have an unpopular opinion, mandates are bad for industry including agriculture.

• Are these tariffs not just causing more planting more acres in Brazil?

• Is free trade dead?

Subscribers can access the slides below.

Read more

I’ve noticed “Bond Vigilantes” being mentioned in both headlines and various newsletters over the past few weeks. The graph above shows the number of times that the term is being searched in Google from 2009 to present. It’s important to understand the history of the term and why it matters now.

Economist Ed Yardeni coined ‘Bond Vigilantes’ in 1983. He used the term to refer to bond market traders who would punish countries by selling their bonds, thus driving up interest rates, if they became worried about nations’ inflation rates, deficits or debts.

I first became acquainted with the term in 1995. I had started as a commodity broker at CA Global Futures AG in Vienna, Austria. We operated a small ‘up desk’ in the very heart of the city, with clients from Austrian banks to the Slovakian national oil company. My responsibilities were primarily in physical commodities (LME copper and aluminum, and IPE & NYMEX crude oil and products) though because we were a small desk, we all were familiar with and brokered nearly every futures contract imaginable—Azuki beans in Japan; Bunds on the DTB and LIFFE; London and New York coffee; and Kansas City, Minneapolis, Chicago, and Paris wheat.

In late 1993, and throughout 1994, interest rates rose for nearly every country around the world. After the collapse of the British Pound and its ejection from the European Monetary System, financial markets had become very skeptical of the ability of nations to operate their finances in preparation for the Euro. Specifically, they were worried that increasing debts and deficits would make European monetary integration impossible. At first, those worries were directed at more profligate nations, but in time those worries spilled over everywhere, including the US and Japan.

The rate increases were sharp, in the US, 10 year interest rates went from 5.36% to 7.96% in 14 months. In Japan, rates climbed from 3.4% to 4.8% in eight months. After the markets of 2020-2023, these changes might not seem so severe, but a key difference is that there were no meaningful changes in these economies leading up to the so-called Great Bond Massacre of 1994. There was no pandemic or rapidly accelerating inflation. There was no rapid acceleration of deficits or debt; the market simply changed its mind about how much it wanted compensated for the risks it was bearing.

Globally, nations moved quickly toward fiscal austerity. Europe went into recession in 1995 and 1996, depending on the country. Why? Because the Bond Market vigilantes changed their minds one morning in late 1993.

I am often asked about the debt, and whether it matters. I like to quote Dick Cheney in 2003, who said, “Reagan proved that deficits don’t matter.” I like to append “until they do” to Cheney’s quip.

If we look at the debt-to-GDP of these same countries, the first thing to notice is that there really isn’t an obvious reason for the Vigilantes to have intervened when they did in 1994, especially when we look at debt levels today. While Italy’s debt level is similar to its level in the early 1990s, Japan’s has tripled, and US and UK levels have doubled, yet we haven’t seen that same sort of spike. At least not yet.

But grumblings are growing on Wall Street. In the 2024 election, neither candidate paid even lip service to fiscal rectitude. Both promised many and varied cuts in taxes and increases in spending. Both seem to buy more into Dick Cheney’s take than mine. History doesn’t look kindly on this silence, and neither do the bond markets. Based on the moves in the past few weeks, I expect to see a continued steepening of the yield curve. The Fed has cut rates again by 0.25% today and I think they are done until they get stronger signals of a slowing economy. The real risk is bond market downside—as worries about growing deficits and debts, and potentially more worries about a resurgence of inflation in response to tariffs and immigration policy.

If you have exposure to longer interest rates, whether through bond investments or borrowing, the risk is moving toward bonds and away from equities. When will it start happening? When the wind shifts, the moon is right, and attitudes change. Maybe in a week, maybe in a year, but the current path risks awakening those vigilantes once again.

Thursday afternoon export figures indicate that corn sales remain on pace to finish up the year by meeting the USDA’s (May) export estimate of 2,150 million bushels, however shipments remain somewhat behind the pace we’d expect for that level of commitments. As of now, commitments are at 93.9%, compared to an average for this time of year of 92.4%, which is a difference of 35m bushels ahead of pace.

From figure 1, shipment pace is a bit slow, leaving outstanding bushels a bit high (both outside of the 25th/75th percentile band) but these are relatively small deviations, and I think point to exports remaining unchanged in next week’s WASDE report.

Soybeans

Read more

Many states (and farmers) worked through a wetter-than-normal late April and early May, enough that concerns grew about planting progress and the eventual impact on 2024 yields. New crop corn prices rallied from $4.60 to $4.95 between April 15th and May 15th.

During mid-May, weather patterns changed, and planting progress accelerated, resulting in prices giving back that rally in the weeks since May 15th.

Using the weather model for corn from this earlier post, I calculated the currently predicted yield by state using the most recent data available. I found two things: first, the planting rate doesn’t indicate any reason to be worried about yield this fall. Second, that while there are some differences between using the May 15 date recommended by FarmDoc and others to measure late planting, that date matters much more for Illinois, Indiana, and Iowa than others.

In the table below are projections of 2024 corn yield by state for the 10 largest corn producing states in the US. The first data column (June 3rd) is the projected yield using all of the data available as of the June 3rd, 2024 Crop Progress report, specifically, lateplanted, SepMarPrecip, and AprTemp for 2024 are used to calculate the projection, though the model is only estimated with data through 2023.

The column Jan 1st is the projection for 2024 yields by state ignoring what we know has happened so far in 2024.

                 2024 Yield            Prospective   Proj. Prod. (m bu)
        State  June 3rd  Jan 1st   se   Plantings   Jun 3rd   Jan 1st
      INDIANA   183.8     186.0   3.65     5100       948       937
     ILLINOIS   197.6     199.0   3.20    10900      2168      2153
         IOWA   202.9     204.0   2.76    12800      2611      2597
    MINNESOTA   196.1     197.2   4.51     7900      1557      1549
     MISSOURI   162.7     156.8   4.53     3500       548       569
     NEBRASKA   189.8     189.5   2.26     9850      1866      1869
 NORTH DAKOTA   137.6     140.0   7.20     3800       532       523
         OHIO   187.2     183.2   4.08     3300       604       617
 SOUTH DAKOTA   165.0     163.3   3.65     6000       979       989
    WISCONSIN   177.6     178.9   3.61     3800       679       674
        Total   186.4     186.7   n/a     66950     12498     12481

As you can see, the differences this year between the Jan 1 and the June 3 models aren’t large, particularly when compared to the standard error of the model, which is an estimate of the variability of the prediction. In general, the only state that so far exhibit more than one standard error of difference is Missouri, though Ohio is within a hair (0.08 standard errors) of doing so. (Is it a coincidence that these are the only two states in this list I’ve lived in?)

When we add the acreage estimates from the March 31st Prospective Plantings report, and then calculate the net effect on production, we get a similar story, with very small differences. In the total bushel calculation, it is also easier to see that the changes largely cancel out, for only a total of 18m bushel greater forecast production for these 10 states.

So, to summarize, from what we know so far, we are still on track for a trend line yield in the US for corn. In future posts, I’ll run the same analysis for soybeans, and present what it means for my balance sheets for both crops.

Just as in my previous post about corn, a quick glance at national average yield for soybeans might lead one to believe that soybean yields have stopped increasing. The last record yield in the US was 2016 at 51.9 bushels/acre. Since then, we’ve exceeded 51 bushels/acre on two more times, but we’ve gone 7 years without a new yield record.

As with corn, the issue here is not breeding, or lack thereof, or traits, but weather. 2016 presented an excellent year for weather conditions for soybeans, and no year since then has matched it.

Subscribe now

We are all aware that there are many factors that affect yield in any given field or farm. But as we aggregate to county, state, and national yields, differences in fertility, drainage, seed, even pest and disease pressure give way to the effects of weather.

Since 2013, the USDA World Agricultural Outlook Board has used a weather-informed model (Westcott and Jewison (2013) ) to project yields. This model has seen many variations applied by other analysts, including notably FarmDoc Daily of the University of Illinois. I used this model to estimate yields for corn for 2024.

                 estimate    s.e.    t-stat   p-value
(Intercept)	-1047.847    58.68   -17.86    <0.001
Year	            0.543     0.03    20.87    <0.001
lateplanted	   -0.040     0.02    -2.13	0.042
SepMarPrecip	    0.095     0.13     0.73	0.470
AprPrecip	    0.215     0.40     0.54	0.590
JunPrecip	    2.639     1.70     1.55	0.131
JunPrecip2	   -0.287     0.18    -1.58	0.124
JulPrecip	    2.943     1.82     1.61	0.117
JulPrecip2	   -0.274     0.20    -1.38	0.179
AugPrecip	    0.834     0.34     2.43	0.021
AprTemp	            0.041     0.11     0.39	0.700
MayTemp	            0.104     0.12     0.88	0.387
JunTemp	            0.471     0.17     2.76	0.010
JulTemp	           -0.402     0.16    -2.46	0.020
AugTemp	           -0.378     0.16    -2.35	0.026

When applied to soybeans, the power of the model declines a bit—it only explains 93.96% of the variation in yields compared to 95.57% as for corn. But 93.96% is nonetheless very good performance.

This model also allows us to estimate a weather-adjusted trend yield for 2024. Using these results, and without any data yet for 2024, the yield forecast for soybeans in the US is 51.8 bushels/acre.

Soybeans also exhibit continued annual average growth in yields—of 0.54 bushels/year. The factors that most significantly affect yields are late planting, August Precipitation, and Temperature in June, July and August. Winter precipitation (“SepMarPrecip”) has no clear statistical impact on yields when the other factors are included.

Over the past five years, we can see that for soybeans, in the years in which actual yield deviation from trend (the white line) were negative—2019 and 2022, late planting played a role in both. In 2019, Summer Precipitation largely offset the planting, while in 2022, it reinforced it.

Going forward, the next question is when does 2024 matter? In the past, evidence of the predictive power of planting progress reports was very limited until late April, and only then became substantial as May progressed. However, as we are now in the period in which planting progress is being released, we can update these models for the present date, and compare to the past to see whether current progress, or lack thereof, is enough to affect expected harvest yields.

In many farmer meetings in late 2022 and early 2023, I heard increasing skepticism about the direction of corn yields. Farmers, agronomists, end economists wondered whether the lack of new traits and the continued expansion of corn and soybean acreage meant that corn yields were destined to plateau in the upper 170s until some new breakthrough came along.

These beliefs are paticularly problematic from a marketing perspective. When the first forecasts each year begin to emerge in the Winter and Spring before planting, doubt around forecasts of yields above 180 means that farmers with those views are likely to see the market as irrationally bearish, hurting their own marketing decisions.

Having worked in commodity analysis since 1995, and being a bit of a data nerd on top of that, this all piqued my curiousity. Have yields actually stopped, or paused growing? What is the effect of acreage expansion into marginal areas? If not in yield growth, have there been changes in yield variability or other characteristics of the trend in recent years? I’ve been working on these questions over the past six months, but some changes delayed my publication of the results. Over the coming weeks, I hope to fix that.

The source of the doubts about yield growth are pretty apparent. In 2014, US National yields first broke 170 bushels/acre, at 171. Since then, seven of the nine years have been betwen 171 and 177 bushels/acre, and graphically, yields do appear to have flattened out since 2016. While we’ve seen new records, even in 2023, they’ve been slight at best.

Starting in 2013, the World Agriculture Outlook Board has informed its yield forecasts with weather, using the model in Westcott and Jewison (2013). Since then, FarmDoc Daily has published a number of additional articles on the interaction of weather and national yields. The use of weather data provides an opportunity to dig deeper into yields over the past decade. Irwin (2023) estimates a relatively simple model of national average yield taking into the account of weather. Using the same weather data from the Midwestern Regional Climate Center and yield data from NASS, and estimating with data from 1980 to 2023 data yields the regression model of:

                 estimate     se.       t-stat	p-value
(Intercept)	-3631.097   130.64	-27.79	<0.001
Year	            1.954     0.06	 31.84	<0.001
lateplanted	   -0.345     0.09	 -3.99	<0.001
SepMarPrecip	    0.406     0.35	  1.16	 0.256
AprPrecip	    2.482     1.08	  2.30	 0.029
JunPrecip	    7.522     4.21	  1.79	 0.084
JunPrecip2	   -0.842     0.44	 -1.92	 0.065
JulPrecip	   31.054     4.45	  6.98	<0.001
JulPrecip2	   -3.330     0.49	 -6.84	<0.001
AugPrecip	    2.545     0.83	  3.07	 0.005
AprTemp	            0.097     0.30	  0.33	 0.744
MayTemp	            0.326     0.31	  1.06	 0.298
JunTemp	           -0.365     0.42	 -0.87	 0.390
JulTemp	           -2.227     0.41	 -5.44	<0.001
AugTemp	           -1.061     0.41	 -2.61	 0.014

Adj R^2: 0.9557

This is very similar to the results that Irwin obtained in the article referenced above. The R^2 (a measure of how well the explanatory variables ‘explain’ the independent variable, which is yield) is very high, and the overall model is very significant. The single most important number here is the estimate for ‘Year’ which represents the estimate of how much US national average corn yield is increasing each year, after adjusting for weather. This model shows an estimate of 1.954 bushels per year, and that estimate is very very statistically significant.

Figure 3 compares actual yields, as estimated by NASS each year from 1980 to 2023, and compares them to the weather-adjusted predicted yields. The blue line uses data from 1980 to 2015, before yields ‘flattened’ and the red line uses data through 2023. What we see from both models of yield is that including weather in the analysis shows that yields should have been lower since 2015. Further, when we compare the predicted yield based on 1980-2015 to that of 1980-2023, we see that given the actual weather in 2016-2023, the red line is higher than the blue, which indicates (doesn’t prove, just indicates) that in fact yields are becoming even more tolerant of adverse weather.

Figure 4 attempts to convey the impact of the different weather conditions on yield in each of the years since 2015 from the regression model above. To simplify the visual, I’ve combined a number of the different variables. We see, for example, that in 2023, on net, summer temperatures were actually a positive for yield, while spring precipitation and late planting were both negatives, and the effects of was largely offsetting. Note that just because the modeled yield in 2023 shows only -0.6 bushel change, there is still an idiosyncratic bit of the model, i.e. it’s not perfect. The very hot and dry conditions in the Western Cornbelt and the nearly perfect conditions in the Eastern Cornbelt aren’t easy to model without diving to the state level, which will be a coming post.

In 2022, the effects of weather were much clearer with large amounts of late planting, as well as negative summer temperatures resulting in yield that was 7 bushels below trend. When we look at the effect of weather across all of the past seven years, we see that not since 2018 would we have expected weather to actually increase yield.

Finally, for the punch line…what happens when we estimate yield for 2024? Well, we don’t yet have precipitation data for September 2023-March 2024, as in the model above—we should have it in the next week or so—but I arrive at an estimate of 182.1 bushels/acre assuming ‘average’ weather.

These are watermarked versions of the graphs that I use in my presentations. Feel free to use them in your own presentations. If you’d like data graphed that isn’t here, or graphed differently, or a slide is problematic, please email me ASAP.

Thanks for subscribing,

matt.

Read more

As steam builds toward decarbonizing the economy, there is understandably more concern and confusion about carbon offset payments and I’ve been fielding increasingly more questions about the carbon market. Over the past few weeks, I’ve learned a few things that I think could help to demystify where the agriculture community could go with payments, as well as how to get there.

On 25 January 2022, I was on a panel with the excellent Lars Dyud, CEO of EarthOptics, at the 2022 Independent Seed Professionals Association annual meeting in Indianapolis. A couple of points I took from his presentation:

1. An acre foot of soil weighs about 2000 tons.1

2. When we speak of sequestering carbon in soil, we are actually referring to raising the organic matter levels in the soil. Roughly 60% of soil organic matter is carbon.2

3. EarthOptics’ technology is really cool and has lots of potential to make carbon levels (and nutrient levels) much more granular across a field and actually verifiable—opening the door to documenting changes in sequestration (payment for results) instead of merely changing how you operate (payment for practices).

With these facts in hand, I started doing some back of the envelope calculations on what the potential value of carbon sequestration could be in terms that I can understand. If a farmer raises the soil organic matter by 1%, that is an additional 20 tons of SOM, of which 60% is carbon, or 12 tons. However, CO2 is 3.67 times heavier than carbon alone3 so that this 12 tons of carbon is equivalent to 44 tons of CO2.

The market price of carbon varies a lot, like the prices for most traded commodities. Historically, we used a price in the US of $50/ton as a benchmark to rein in GHG emissions. Sequestering farmers wouldn’t receive the full market price. There will be some significant costs for verification, for example, but as the market is currently trading above €90/tonne in Europe, let’s just say we leave lots of space for transactions costs and assume $25/ton. At this price, the potential value of sequestered carbon is $1,100/acre for every 1% increase in soil organic matter.

I’m told that raising SOM by 1% is a process that might take 10-20 years, converting that $1,100/acre into $55-$110/acre per year—a far cry from what’s currently being offered to farmers. Why the difference? Lack of verification? Payment for practices vs. results? Lack of federal limits on carbon? Yes.

So until those things change, per-acre payments will remain much lower than they could be. Is there potential here? Also yes.

Apple and Google have released data drawn from cell phone usage as proxies for compliance with stay-at-home orders. The Apple data is a measure of the change in requests for directions to Apple Maps, by driving, transit, and on foot. The Google data reports the relative change in visits to retail, grocery/pharmacy, parks, etc. Helpfully, they have been combined, and their data format has been standardized by GitHub user ActiveConclusion.

These datasets came to mind recently when I was looking at the recovery in gasoline consumption in the US as stay-at-home restrictions have been relaxed (see plot above). I was curious what the relationship between the Mobility data and Gasoline consumption would actually be. Obviously, we know that there is a connection, but how complete is the mobility data? In order to find out, I ran a simple regression model of Total US Driving Index from Apple on Motor Gasoline supplied from the EIA. (full code and results below)

Fitting a model to the two series (I’m just exploring here, so don’t anyone get wound up about the lack of sophistication of this model) indicates a very strong relationship–changes in the mobility index explain 84.2% of the changes in gasoline supplied on a weekly basis, and each one point change in the Mobility Index results in a 0.8% change in gasoline usage. These results are also very highly statistically significant. What is the practical implication? As Apple is releasing its data each day, whereas the EIA data is released only weekly, with a five day lag (each Wednesday’s release is for the week ended the previous Friday). Using this quite simple model, it would permit an analyst to create potentially more accurate forecasts of gasoline consumption ahead of the EIA release each week.

Regression Results for the model ln(gasoline) = b_0 + b_1 * driving + e

Some of the biggest news of the week was the release of the Consumer Price Index for April on May 12. The headline news was that the Consumer Price Index for All Urban Consumers fell 0.8% in April, the largest decline since December 2008, driven largely by the 20.6% fall in gasoline prices. If we exclude food and energy prices, the fall was still 0.4%, and was driven by 4.7% declines in both airline fares and apparel prices.

All food prices were up by 1.5%, led by food at home prices rising by 2.6%, compared with food away from home rising at only 0.1%. Food at home categories nearly all showed higher prices, with Meat, Poultry and Fish rising at 3.6% as a whole, and nearly every category rising with it, paced by Other Pork, including Roasts, Steaks & Ribs rising at 10.1%. Cereals and Bakery products were another strong category, with prices rising 2.9%, paced by Rice (3.6%), Fresh Biscuits, Rolls & Muffins at 4.7% and Cookies at 5.1%. Finally, eggs were another big riser at 16.1% from March to April.

April was the first month in which the majority of the country was locked down, but it was only in the latter half of the month that wide-spread closures of meat-packing plants started. We are just seeing the effects of the COVID pandemic in most of the economic data this month. For price changes, the impact is obvious in both food and energy, but is not widely obvious in the ‘core’ CPI data, outside of airline prices (for obvious reasons) and apparel, which is heavily driven by Men’s suits, sport coats, and outerwear, and women’s dresses categories.

The May numbers next month will show higher increases in food prices, compounded by a rise in gasoline prices. The more interesting question is always how much effect we will see in core consumption. With consumer spending only starting to recover in May, as the country reopens, price changes will be an early indicator of the resiliency of the US consumer.

• [Not the Great Depression] The first thing to get out of the way is that I don’t think we are entering a second Great Depression. Having said that, we will likely see unemployment levels much greater than what we saw during the Great Recession, and some of the quarterly GDP and unemployment statistics may be similar to what was observed in the Great Depression.

The Great Depression lasted from 1929 until the beginning of World War II. At its worst, unemployment in the US was 24.9%, and GDP declined by 26.3% (1932 and 1933). These numbers are devastating. For context, in December 1982, the unemployment rate hit 10.8%; it reached a maximum of 10.0% in October 2009, and outside of those two instances, it hasn’t exceeded 9% since the Great Depression.

• [Not the Great Recession] The great recession was, at its heart, a banking crisis. The Great Depression had many causes, but it also precipitated a banking crisis. I do not expect to see a banking crisis associated with the current situation. Banks are much better capitalized and regulated than they were 12 years ago. Further, though I have many problems with the ‘Too Big to Fail’ designation, it is a public recognition of how damaging financial crises are. It is generally accepted that recessions that have financial crises are slower to recover, and therefore more severe, than those that are not. Watching the fortunes of the banking sector will be a very important indicator of the severity of this crisis.

• [Similar Severity to the Great Recession] While it will be different than the Great Recession, as I don’t foresee a banking crisis, I think that it will be remembered as similar in overall effect. Likely the best analog is the Financial Panic of 1907, which, although it was a financial crisis, and a very, very serious and acute one, it was very quickly managed through the efforts of JP Morgan. Within 18-24 months, the nation had largely recovered. This is what I believe the current crisis will most resemble–a sharp fall in consumer demand that reverberates through the economy, but the economic damage is narrower than in many previous recessions, and is largely driven by consumers permission to spend, not their ability or willingness. As that permission is restored, the economy can start to heal.

• [What Else?] The Saudi-Russian oil war of early 2020 is contributing to the decline in the US economy. Only 15 years ago, such a situation would be stimulative to the US economy, as we were much larger consumers than producers of energy. The shale revolution has changed that equation to turn the US into the largest oil producer in the world. We are also by far the largest producer that does not have the explicit ability to set production. If Saudi Arabia, Russia, or Venezuela wants to cut production in order to manage global output, they can do so. The US, with the exception of the Texas Railroad Commission for Texas producers, cannot. Therefore, any production cutbacks in the US will have to be forced through the price mechanism, which, as we’ve seen, is brutal.

• [Is there more?] Of course. This also assumes that we don’t descend into some sort of trade war. It is not a coincidence that Smoot-Hawley was passed during the depths of the Great Depression, and we are hearing many calls for a further reduction in US trade, or for an ‘on-shoring’ of US supply chains. If forced by revised and/or revoked trade deals, such actions will only prolong and deepen the current crisis.

In the coming days and weeks, I’ll be posting more commentary on different sectors of the general and agricultural economies. But since I had to put this down in writing for a colleague (okay, my wife/business partner), it made sense to put it up here, too.

Leaving OSU was always going to be an adventure, but one of the things that I really looked forward to was the freedom to learn and teach about a wider variety of topics than I might’ve at OSU. Password security was the first product. But now, I’d like to introduce you to my next, and much more in-depth project: Robo-advising.

What are Robo-advisors? Robo-advising is the natural outgrowth of rising investor awareness of the importance of minimizing fees on their portfolio, ever-cheaper computing power, and investors’ increased comfort with online relationships for everything from shopping to medical advice to investing. Robo-advisors are investment accounts who’s investment strategy and transactions are performed automatically and algorithmically based on the offering firm’s strategy and assessment of the investor’s goals and risk appetite.

So this is just a review, right? Yes and no. When I started at Ohio State in 2001, we had the choice between joining the traditional state pension system or the Alternative Retirement Program (ARP). Enrollees in the ARP opened 403(b) accounts (similar to a 401(k)) and all of the contributions that would normally have been made to the pension system were instead deposited into our 403(b) accounts. The kicker is that the state of Ohio has never joined the Social Security system, so the contributions to the 403(b) had to be at least as large as our SSI contributions–employer and employee. When I left OSU, I rolled all of this into a traditional IRA with Charles Schwab–where I opened my first online trading account in 1986–to escape the high fees associated with the 403(b) account I was in.

What about the robo-advisors? I’d read and heard a lot about Robo-advising, and was intensely curious. I came to the realization a number of years ago that the best thing I could do to improve my retirement prospects was to spend less time ‘investing’ and instead focus on earning and saving. Over the past 15 years, I looked at my investment account balances about once per quarter, rebalancing them when I felt like it/remembered to, and changing investment allocations about every three years. The prospect of an account that could ostensibly do all of that for me is pretty attractive.

Which Robo-Advisor? Who knows? Betterment and Wealthfront seem to be the original robo-advisors, or at least the firms that really popularized the concept, but now, nearly every investment firm, large and small, seem to offer some sort of robo-advising platform, including [Schwab][intelligent.schwab.com]. But reading dozens of articles made it very clear that not much is clear about the different platforms. They are all slightly different, and of course, the algorithms that make allocation recommendations are also different. However, I could narrow it down to three:

• Betterment is one of the original robo-advisors, and clearly bends over backwards to market to young savers. It has low fees, and is 100% online only.

• Charles Schwab Intelligent Portfolios is the name of Schwab’s robo-advisor service. I’ve been a Schwab customer for nearly 2/3’s of my life, and I’ve never been unhappy with them. They do nearly everything, and though they may not be best-of-breed, Schwab is always competent.

• Liftoff is the outlier here. When I’ve told others what I’m doing with this project, and mentioned Betterment and Schwab, I got knowing nods, but Liftoff? Who’s that? Liftoff is the robo-advisor of Ritholtz Wealth Management. RWM’s namesake Barry Ritholtz is a contributor to Bloomberg, and produces one of the best financial podcasts out there, Master’s in Business. But it’s really Josh Brown’s Reformed Brokerblog that first introduced me to RWM. I’ll admit that I have a bit of an intellectual/professional man-crush on Josh. His writing is excellent, his insights profound, and his wit is cutting. Finally, as a reformed broker myself, there is some small bit of kinship I feel.

Because I didn’t feel like I could narrow it down any further without actually being a client, I just opened accounts at all three. I wouldn’t buy a car without test-driving it, so why would I trust the largest asset on my balance sheet to someone cold-turkey? Over the coming year, I’ll be updating this series with information on my experiences with the three.

The picture above encapsulates a lot about my previously-discussed transition from Ohio State University to Kernmantle Group. For the past 15 years, all of personal computing has been done on OSU computers, and the majority of them have been MacBooks.

On the left is a 13” MacBook Retina (early 2015), i5, 8GB RAM, and 512GB SSD, which cost $2200. In front of it is a 12.9” iPad Pro 32GB, which with the Apple Pencil, costs $850, for a total of $3050. This has been my OSU setup since the iPad Pro was released, and this laptop, and a 13” Retina i7 prior, have been my steady travelling buddies for the last 5 years. While $3k seems like a lot, honestly, its not that much when it’s someone else’s money. And while that sounds pretty callous, it is entirely true.

However, last December, I left OSU and so had to replace my computers. In doing so, I learned a few things, but before I get to that, let’s talk about how got to here.

To replace my OSU setup in moving to Kernmantle, I needed a few things:

• Sufficient disk space and computing power for everything from crunching lots of data to editing speaking videos, and the ability to drive at least 2 large (WQHD–2560x1440) monitors.

• On the road, a laptop with enough power to run RStudio to update my charts and plots, and do it on the machine, not via RD back to my home office.

• Pen-based input. For me, the iPad Pro has been excellent. I had a Dell 2-in-1 about 10 years ago which was a so-so laptop but it really made me love pen-based note-taking. In 2007, I travelled to South Africa for 10 days on a research project, during which time I conducted a dozen or so interviews. In OneNote, I could quickly write notes while also recording the audio for later review.

• Small and light. I already travelled a lot, and with Kernmantle, I’m going to be travelling significantly more. While the 13” Retina + iPad Pro aren’t large, combined with their chargers, I can feel the weight in my briefcase/backpack.

Step 1: Get a desktop for the home office.

I realized pretty quickly that the best way to get all the power and disk space at home was a desktop Mac. However, the new MacPros are entirely prohibitive. They start at $2699 and don’t have the storage space that I’d like to have, so it would be at least $3000 to get the computer that I’d need. The next option was to build a Hackintosh. But as this was going to be my main computer, the inevitable tinkering was a definite worry. In researching how to build a Hackintosh, I found that many of the tips and tricks came from users with older MacPros. Apparently, many professionals still use the ‘cheese-grater’ MacPro models that were discontinued in 2012. Those made from 2009 to 2013 can still run the latest operating system, they can hold up to 4 internal drives and they are highly expandable for video cards, etc. After some looking on eBay, I ended up with a Quad 3.2GHz Xeon machine for $400. I added a 512GB SSD and two 3TB hard drives and had a fast, capable machine that could easily run a 1080p monitor as well as my new ASUS WQHD monitor.

Next time, I’ll discuss the hardest part: the travelling office.