diff --git a/python/pyarrow/array.pxi b/python/pyarrow/array.pxi
index 0d6af9eebb73d..b8137c9e04b4e 100644
--- a/python/pyarrow/array.pxi
+++ b/python/pyarrow/array.pxi
@@ -1682,10 +1682,8 @@ cdef _array_like_to_pandas(obj, options, types_mapper):
     original_type = obj.type
     name = obj._name
 
-    # ARROW-33321 reenables support for date/timestamp conversion in pandas >= 2.0
-    from pyarrow.vendored.version import Version
-    if pandas_api.loose_version < Version('2.0.0'):
-        # ARROW-3789(wesm): Convert date/timestamp types to datetime64[ns]
+    if pandas_api.is_v1():
+        # ARROW-3789: Convert date/timestamp types to datetime64[ns]
         c_options.coerce_temporal_nanoseconds = True
 
     if isinstance(obj, Array):
diff --git a/python/pyarrow/pandas-shim.pxi b/python/pyarrow/pandas-shim.pxi
index 7dc5d590a72af..8a27173a00d91 100644
--- a/python/pyarrow/pandas-shim.pxi
+++ b/python/pyarrow/pandas-shim.pxi
@@ -58,6 +58,7 @@ cdef class _PandasAPIShim(object):
         self._pd = pd
         self._version = pd.__version__
         self._loose_version = Version(pd.__version__)
+        self._is_v1 = False
 
         if self._loose_version < Version('1.0.0'):
             self._have_pandas = False
@@ -72,6 +73,8 @@ cdef class _PandasAPIShim(object):
                     "installed. Therefore, pandas-specific integration is not "
                     "used.".format(self._version), stacklevel=2)
                 return
+        elif self._loose_version < Version('2.0.0'):
+            self._is_v1 = True
 
         self._compat_module = pdcompat
         self._data_frame = pd.DataFrame
@@ -150,6 +153,10 @@ cdef class _PandasAPIShim(object):
         self._check_import()
         return self._version
 
+    def is_v1(self):
+        self._check_import()
+        return self._is_v1
+
     @property
     def categorical_type(self):
         self._check_import()
diff --git a/python/pyarrow/pandas_compat.py b/python/pyarrow/pandas_compat.py
index e8d89b82eace3..630cdb9328960 100644
--- a/python/pyarrow/pandas_compat.py
+++ b/python/pyarrow/pandas_compat.py
@@ -787,10 +787,8 @@ def _reconstruct_block(item, columns=None, extension_columns=None):
 
 
 def make_datetimetz(unit, tz):
-    from pyarrow.vendored.version import Version
-    # ARROW-3789: Convert date/timestamp types to datetime64[ns]
-    if _pandas_api.loose_version < Version('2.0.0'):
-        unit = 'ns'
+    if _pandas_api.is_v1():
+        unit = 'ns'  # ARROW-3789: Convert date/timestamp types to datetime64[ns]
     tz = pa.lib.string_to_tzinfo(tz)
     return _pandas_api.datetimetz_type(unit, tz=tz)
 
diff --git a/python/pyarrow/table.pxi b/python/pyarrow/table.pxi
index 794918d80dd2b..4ed50f191e952 100644
--- a/python/pyarrow/table.pxi
+++ b/python/pyarrow/table.pxi
@@ -2892,10 +2892,8 @@ def table_to_blocks(options, Table table, categories, extension_columns):
         c_options.extension_columns = {tobytes(col)
                                        for col in extension_columns}
 
-    # ARROW-33321 reenables support for date/timestamp conversion in pandas >= 2.0
-    from pyarrow.vendored.version import Version
-    if pandas_api.loose_version < Version('2.0.0'):
-        # ARROW-3789(wesm); Convert date/timestamp types to datetime64[ns]
+    if pandas_api.is_v1():
+        # ARROW-3789: Convert date/timestamp types to datetime64[ns]
         c_options.coerce_temporal_nanoseconds = True
 
     if c_options.self_destruct:
diff --git a/python/pyarrow/types.pxi b/python/pyarrow/types.pxi
index d0efbe547eb4e..f6baa4ae40e83 100644
--- a/python/pyarrow/types.pxi
+++ b/python/pyarrow/types.pxi
@@ -42,7 +42,8 @@ cdef dict _pandas_type_map = {
     _Type_HALF_FLOAT: np.float16,
     _Type_FLOAT: np.float32,
     _Type_DOUBLE: np.float64,
-    _Type_DATE32: np.dtype('datetime64[D]'),
+    # Pandas does not support [D]ay, so default to [s]econd for date32
+    _Type_DATE32: np.dtype('datetime64[s]'),
     _Type_DATE64: np.dtype('datetime64[ms]'),
     _Type_TIMESTAMP: {
         's': np.dtype('datetime64[s]'),
@@ -131,9 +132,8 @@ def _get_pandas_type(type, unit=None):
     if type not in _pandas_type_map:
         return None
     if type in [_Type_DATE32, _Type_DATE64, _Type_TIMESTAMP, _Type_DURATION]:
-        from pyarrow.vendored.version import Version
-        # ARROW-3789: Convert date/timestamp types to datetime64[ns]
-        if _pandas_api.loose_version < Version('2.0.0'):
+        if _pandas_api.is_v1():
+            # ARROW-3789: Convert date/timestamp types to datetime64[ns]
             if type == _Type_DURATION:
                 return np.dtype('timedelta64[ns]')
             return np.dtype('datetime64[ns]')
@@ -1161,7 +1161,7 @@ cdef class DurationType(DataType):
         >>> d.to_pandas_dtype()
         timedelta64[ms]
         """
-        return _get_pandas_type(_Type_TIMESTAMP, self.unit)
+        return _get_pandas_type(_Type_DURATION, self.unit)
 
 
 cdef class FixedSizeBinaryType(DataType):