From the Purdue/CME Ag Econ Barometer:
Figure 1: Ag Econ Barometer Index (blue, left scale), Consumer Confidence (brown, right scale). The NBER has defined recession days as shaded in gray. Ag Econ Barometer Oct 2015-Mar 2016 = 100. Source: Purdue/CME, Conference Board, NBER.
The trade war has continued in earnest for over a year – since “Liberation Day”, and now there are two months of military hostilities with Iran. Despite the well-publicized hits to farm incomes caused by the trade war, the barometer remains relatively high – beyond the end of 2024. This is an expression of action Economist commented on “Donald Trump is crushing America’s farmers – yet they support him”
What (empirically) explains farm sentiment? One may be tempted to look at the high prices of agricultural commodities compared to input costs. In Figure 2, I plot the Barometer against the log of agricultural commodity prices and fertilizer prices:
Figure 2: Ag Econ Barometer index (blue, left scale), log average of world value of agricultural commodities over fertilizer (red, right scale). The NBER has defined recession days as shaded in gray. Orange shading denotes the Trump administration. Source: Purdue/CME, World Bank, NBER, and author’s calculations.
Interestingly, Barometer declines in agricultural commodity prices and fertilizer prices yield an adjusted R.2 Only 7%. Note, obviously, a large number of differences are explained by the Trump administration. (AGBAROMETER it strongly rejects the common unit root test, and fails to reject the KPSS trend test).
AGBAROMETER = 2.62 + 0.29TRUMP + 1.61AGPRICE – 0.19FERTPRICE
Adj-R2 = 0.34, SER = 0.17, DW = 0.66. Data for 2016M01-2026M04. Bold coefficients indicate significance at the 10% level using robust HAC standard errors.
AGBAROMETER is an index of 100 points divided by 100 percent, AGPRICE is the real log price (main CPI reduced) of agricultural goods, ad FERTPRICE is the corresponding reduced log price of fertilizers.
It is difficult to see the relative influence of these variables, so I report the coefficients (standard deviations) below:
AGBAROMETER = 0.70TRUMP + 0.63AGPRICE – 0.31FERTPRICE
In other words, one standard deviation of the TRUMP variance also increases the AGBAROMETER by 0.7 standard deviations.
One would think that support for farmers would decrease. The Economist article says:
It would be easy to blame Mr Trump for the collapse; after all, he campaigned with promises to lower prices and renew the heart. But rural America does not. The president’s approval ratings are higher among rural voters than any other group in our survey. Most still think he is doing a good job. In an interview with The Economist, farmers said they trust the administration – but they need help to recover the losses their foreign policy is causing them.
The sentiment measured by the Ag Econ Barometer is about the economic outlook farmers are facing. It does not directly measure whether farmers think things are going well. Recently, they started asking the “right way/wrong way” question. Here are the differences, plotted against expectations.
Figure 3: Ag Econ Barometer Expectations index (light blue, left scale), country’s negative direction difference, % (purple, right scale). Ag Econ Barometer Oct 2015-Mar 2016 = 100. Orange shading denotes the Trump administration. Source: Purdue/CME, author’s calculations.
So, while the “right/wrong” difference is still good at 14 points, it actually decreases the longer (shorter) the data we have.
