diff --git a/flox/core.py b/flox/core.py
index 1437e506c..9ff99b583 100644
--- a/flox/core.py
+++ b/flox/core.py
@@ -1344,17 +1344,8 @@ def dask_groupby_agg(
         # find number of groups in each chunk, this is needed for output chunks
         # along the reduced axis
         slices = slices_from_chunks(tuple(array.chunks[ax] for ax in axis))
-        if expected_groups is None:
-            groups_in_block = tuple(_unique(by_input[slc]) for slc in slices)
-        else:
-            # For cohorts, we could be indexing a block with groups that
-            # are not in the cohort (usually for nD `by`)
-            # Only keep the expected groups.
-            groups_in_block = tuple(
-                np.intersect1d(by_input[slc], expected_groups) for slc in slices
-            )
+        groups_in_block = tuple(_unique(by_input[slc]) for slc in slices)
         groups = (np.concatenate(groups_in_block),)
-
         ngroups_per_block = tuple(len(grp) for grp in groups_in_block)
         group_chunks = (ngroups_per_block,)