diff --git a/doc/source/whatsnew/v1.2.2.rst b/doc/source/whatsnew/v1.2.2.rst
index 240acf787f9c9..0ee1abaa2a0eb 100644
--- a/doc/source/whatsnew/v1.2.2.rst
+++ b/doc/source/whatsnew/v1.2.2.rst
@@ -17,6 +17,7 @@ Fixed regressions
 
 - Fixed regression in :func:`read_excel` that caused it to raise ``AttributeError`` when checking version of older xlrd versions (:issue:`38955`)
 - Fixed regression in :class:`DataFrame` constructor reordering element when construction from datetime ndarray with dtype not ``"datetime64[ns]"`` (:issue:`39422`)
+- Fixed regression in :class:`DataFrame.astype` and :class:`Series.astype` not casting to bytes dtype (:issue:`39474`)
 - Fixed regression in :meth:`~DataFrame.to_pickle` failing to create bz2/xz compressed pickle files with ``protocol=5`` (:issue:`39002`)
 - Fixed regression in :func:`pandas.testing.assert_series_equal` and :func:`pandas.testing.assert_frame_equal` always raising ``AssertionError`` when comparing extension dtypes (:issue:`39410`)
 - Fixed regression in :meth:`~DataFrame.to_csv` opening ``codecs.StreamWriter`` in binary mode instead of in text mode and ignoring user-provided ``mode`` (:issue:`39247`)
diff --git a/pandas/core/dtypes/cast.py b/pandas/core/dtypes/cast.py
index c77991ced3907..1338d22f74347 100644
--- a/pandas/core/dtypes/cast.py
+++ b/pandas/core/dtypes/cast.py
@@ -1137,7 +1137,7 @@ def soft_convert_objects(
         # bound of nanosecond-resolution 64-bit integers.
         try:
             values = lib.maybe_convert_objects(values, convert_datetime=True)
-        except OutOfBoundsDatetime:
+        except (OutOfBoundsDatetime, ValueError):
             pass
 
     if timedelta and is_object_dtype(values.dtype):
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 32aade97c8736..99218cebc37e1 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -2503,7 +2503,7 @@ class ObjectBlock(Block):
     _can_hold_na = True
 
     def _maybe_coerce_values(self, values):
-        if issubclass(values.dtype.type, (str, bytes)):
+        if issubclass(values.dtype.type, str):
             values = np.array(values, dtype=object)
         return values
 
diff --git a/pandas/tests/frame/methods/test_astype.py b/pandas/tests/frame/methods/test_astype.py
index d79969eac0323..5264d4b432d34 100644
--- a/pandas/tests/frame/methods/test_astype.py
+++ b/pandas/tests/frame/methods/test_astype.py
@@ -611,3 +611,8 @@ def test_astype_tz_object_conversion(self, tz):
         # do real test: object dtype to a specified tz, different from construction tz.
         result = result.astype({"tz": "datetime64[ns, Europe/London]"})
         tm.assert_frame_equal(result, expected)
+
+    def test_astype_bytes(self):
+        # GH#39474
+        result = DataFrame(["foo", "bar", "baz"]).astype(bytes)
+        assert result.dtypes[0] == np.dtype("S3")
diff --git a/pandas/tests/series/methods/test_astype.py b/pandas/tests/series/methods/test_astype.py
index 3cd9d52f8e754..a3624162a08d8 100644
--- a/pandas/tests/series/methods/test_astype.py
+++ b/pandas/tests/series/methods/test_astype.py
@@ -337,6 +337,11 @@ def test_astype_unicode(self):
             reload(sys)
             sys.setdefaultencoding(former_encoding)
 
+    def test_astype_bytes(self):
+        # GH#39474
+        result = Series(["foo", "bar", "baz"]).astype(bytes)
+        assert result.dtypes == np.dtype("S3")
+
 
 class TestAstypeCategorical:
     def test_astype_categorical_invalid_conversions(self):