BUG: inconsistency in dtype of replace()

doc/source/whatsnew/v1.4.0.rst

@@ -829,7 +829,7 @@ Reshaping
 - Bug in :meth:`DataFrame.stack` with ``ExtensionDtype`` columns incorrectly raising (:issue:`43561`)
 - Bug in :meth:`Series.unstack` with object doing unwanted type inference on resulting columns (:issue:`44595`)
 - Bug in :class:`MultiIndex` failing join operations with overlapping ``IntervalIndex`` levels (:issue:`44096`)
--
+- Bug in :meth:`DataFrame.replace` and :meth:`Series.replace` results is different ``dtype`` based on ``regex`` parameter (:issue:`44864`)
 
 Sparse
 ^^^^^^

pandas/core/internals/blocks.py

@@ -737,7 +737,7 @@ def _replace_regex(
         replace_regex(new_values, rx, value, mask)
 
         block = self.make_block(new_values)
-        return [block]
+        return block.convert(numeric=False, copy=False)
 
     @final
     def replace_list(

pandas/tests/frame/methods/test_replace.py

@@ -1508,3 +1508,16 @@ def test_regex_replace_scalar(
 
         expected.loc[expected["a"] == ".", "a"] = expected_replace_val
         tm.assert_frame_equal(result, expected)
+
+    @pytest.mark.parametrize("regex", [False, True])
+    def test_replace_regex_dtype_frame(self, regex):
+        # GH-48644
+        df1 = DataFrame({"A": ["0"], "B": ["0"]})
+        expected_df1 = DataFrame({"A": [1], "B": [1]})
+        result_df1 = df1.replace(to_replace="0", value=1, regex=regex)
+        tm.assert_frame_equal(result_df1, expected_df1)
+
+        df2 = DataFrame({"A": ["0"], "B": ["1"]})
+        expected_df2 = DataFrame({"A": [1], "B": ["1"]})
+        result_df2 = df2.replace(to_replace="0", value=1, regex=regex)
+        tm.assert_frame_equal(result_df2, expected_df2)

pandas/tests/series/methods/test_replace.py

@@ -604,3 +604,11 @@ def test_replace_nullable_numeric(self):
         assert ints.replace({1: 9.0}).dtype == ints.dtype
         assert ints.replace(1, 9.0).dtype == ints.dtype
         # FIXME: ints.replace({1: 9.5}) raises bc of incorrect _can_hold_element
+
+    @pytest.mark.parametrize("regex", [False, True])
+    def test_replace_regex_dtype_series(self, regex):
+        # GH-48644
+        series = pd.Series(["0"])
+        expected = pd.Series([1])
+        result = series.replace(to_replace="0", value=1, regex=regex)
+        tm.assert_series_equal(result, expected)