It is never too early to begin planning for next season. First, the likely pandemic disruption of fiber market demand from restricted workers/consumers and economic contraction will likely carry into the 21/22 marketing year. Nobody knows, but it strikes me as a long term thing.
The new crop supply question is more immediate. The 21/22 carry-in could start things off heavy by adding over 7 million bales to the U.S. new crop supply. Then there is the planted acreage question. Relative prices (Dec’21 CBOT corn/Dec’21 ICE cotton with a ratio of 5.6 as of October 19) suggest around 13 million acres of U.S. all cotton, based on history (see below).
A more southwestern-specific version of the above chart involves the historical pattern of increasing cotton acres at the expense of hard red winter wheat (see below). This is mostly a consideration in northwestern Texas, Oklahoma, and Kansas. The chart below shows that when the price of wheat, relative to cotton, gets high enough, it is associated with fewer cotton acres being planted. The current ratio of KC Jul’21/ICE Dec’21 cotton prices is 8.2 (as of October 19) which suggests between 6~7M acres of cotton planted in KS-OK-TX. That simply suggests that rising hard red winter wheat futures at present levels might lower southwestern cotton plantings year over year.
The next question concerns moisture and growing conditions. NOAA is forecasting even odds for La Niña conditions in the fall of 2020, which could make it a drier than normal winter. That is on top of the currently dry conditions in most of the southern plains region. So we might expect above average abandonment in the spring of 2021. Dry conditions often loop back to influence above average cotton plantings in the southern plains region.
So, assuming 12.5 million planted acres, historically high U.S. abandonment (29%), and average yield (850 pounds per acre), we can pencil out between 15 to 16 million bale crop and perhaps a 22 to 23 million bale supply. Assuming we either export or spin a total of 18 million bales, that leaves around 4 or 5 million bales of ending stocks for the 21/22 marketing year. This represents a decent year over year reduction in ending stocks. So we might expect somewhat better prices in 21/22 compared to the 20/21 marketing year, all other things equal.