BUG: MultiIndex dtypes incorrect if level names not unique (#45175)

Author: johnzangwill

doc/source/whatsnew/v1.4.0.rst

@@ -838,6 +838,7 @@ MultiIndex
 - Bug in :meth:`MultiIndex.get_loc` raising ``TypeError`` instead of ``KeyError`` on nested tuple (:issue:`42440`)
 - Bug in :meth:`MultiIndex.union` setting wrong ``sortorder`` causing errors in subsequent indexing operations with slices (:issue:`44752`)
 - Bug in :meth:`MultiIndex.putmask` where the other value was also a :class:`MultiIndex` (:issue:`43212`)
+- Bug in :meth:`MultiIndex.dtypes` duplicate level names returned only one dtype per name (:issue:`45174`)
 -
 
 I/O

pandas/core/indexes/multi.py

@@ -739,9 +739,7 @@ def dtypes(self) -> Series:
         from pandas import Series
 
         names = com.fill_missing_names([level.name for level in self.levels])
-        return Series(
-            {names[idx]: level.dtype for idx, level in enumerate(self.levels)}
-        )
+        return Series([level.dtype for level in self.levels], index=names)
 
     def __len__(self) -> int:
         return len(self.codes[0])

pandas/tests/indexes/multi/test_get_set.py

@@ -67,6 +67,23 @@ def test_get_dtypes_no_level_name():
     tm.assert_series_equal(expected, idx_multitype.dtypes)
 
 
+def test_get_dtypes_duplicate_level_names():
+    # Test MultiIndex.dtypes with non-unique level names (# GH45174)
+    result = MultiIndex.from_product(
+        [
+            [1, 2, 3],
+            ["a", "b", "c"],
+            pd.date_range("20200101", periods=2, tz="UTC"),
+        ],
+        names=["A", "A", "A"],
+    ).dtypes
+    expected = pd.Series(
+        [np.dtype("int64"), np.dtype("O"), DatetimeTZDtype(tz="utc")],
+        index=["A", "A", "A"],
+    )
+    tm.assert_series_equal(result, expected)
+
+
 def test_get_level_number_out_of_bounds(multiindex_dataframe_random_data):
     frame = multiindex_dataframe_random_data