34297 sub slow in first call

asv_bench/benchmarks/arithmetic.py

@@ -469,4 +469,29 @@ def time_apply_index(self, offset):
         offset.apply_index(self.rng)
 
 
+class Sub:
+    def setup(self):
+        date_range = pd.date_range("20200101 00:00", "20200102 0:00", freq="S")
+        level_0_names = [str(i) for i in range(30)]
+
+        index = pd.MultiIndex.from_product([level_0_names, date_range])
+        column_names = ["col_1", "col_2"]
+
+        self.df = pd.DataFrame(
+            np.random.rand(len(index), 2),
+            index=index,
+            columns=column_names
+        )
+
+        self.sub_df = pd.DataFrame(
+            np.random.randint(1, 10, (len(level_0_names), 2)),
+            index=level_0_names,
+            columns=column_names,
+        )
+
+    def subsequent_sub_application(self):
+        self.df.sub(self.sub_df, level=0)
+        self.df.sub(self.sub_df, level=0)
+
+
 from .pandas_vb_common import setup  # noqa: F401 isort:skip

pandas/core/indexes/multi.py

@@ -3218,6 +3218,8 @@ def equals(self, other) -> bool:
                 # other cannot contain tuples, so cannot match self
                 return False
 
+            if len(self) != len(other):
+                return False
             return array_equivalent(self._values, other._values)
 
         if self.nlevels != other.nlevels: