BUG: read_csv raising IndexError with multiple header cols, specified index_col and no data rows

doc/source/whatsnew/v1.2.0.rst

@@ -726,9 +726,10 @@ I/O
 - :meth:`DataFrame.to_excel`, :meth:`Series.to_excel`, :meth:`DataFrame.to_markdown`, and :meth:`Series.to_markdown` now support writing to fsspec URLs such as S3 and Google Cloud Storage (:issue:`33987`)
 - Bug in :func:`read_fwf` with ``skip_blank_lines=True`` was not skipping blank lines (:issue:`37758`)
 - Parse missing values using :func:`read_json` with ``dtype=False`` to ``NaN`` instead of ``None`` (:issue:`28501`)
-- :meth:`read_fwf` was inferring compression with ``compression=None`` which was not consistent with the other :meth:``read_*`` functions (:issue:`37909`)
+- :meth:`read_fwf` was inferring compression with ``compression=None`` which was not consistent with the other :meth:`read_*` functions (:issue:`37909`)
 - :meth:`DataFrame.to_html` was ignoring ``formatters`` argument for ``ExtensionDtype`` columns (:issue:`36525`)
 - Bumped minimum xarray version to 0.12.3 to avoid reference to the removed ``Panel`` class (:issue:`27101`)
+- Bug in :func:`read_csv` raising ``IndexError`` with multiple header columns and ``index_col`` specified when file has no data rows (:issue:`38292`)
 
 Period
 ^^^^^^

pandas/io/parsers.py

@@ -1465,9 +1465,14 @@ def _extract_multi_indexer_columns(
 
         # clean the index_names
         index_names = header.pop(-1)
-        index_names, names, index_col = _clean_index_names(
-            index_names, self.index_col, self.unnamed_cols
-        )
+        if not index_names:
+            # In case of no rows and multiindex columns we have to set index_names to
+            # list of Nones GH#38292
+            index_names = [None] * len(ic)
+        else:
+            index_names, _, _ = _clean_index_names(
+                index_names, self.index_col, self.unnamed_cols
+            )
 
         # extract the columns
         field_count = len(header[0])

pandas/tests/io/parser/test_index_col.py

@@ -222,3 +222,18 @@ def test_index_col_large_csv(all_parsers):
         result = parser.read_csv(path, index_col=[0])
 
     tm.assert_frame_equal(result, df.set_index("a"))
+
+
+def test_index_col_multiindex_columns_no_data(all_parsers):
+    # GH#38292
+    parser = all_parsers
+    result = parser.read_csv(
+        StringIO("a0,a1,a2\nb0,b1,b2\n"), header=[0, 1], index_col=0
+    )
+    expected = DataFrame(
+        [],
+        columns=MultiIndex.from_arrays(
+            [["a1", "a2"], ["b1", "b2"]], names=["a0", "b0"]
+        ),
+    )
+    tm.assert_frame_equal(result, expected)