diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 1b07a5b97806e..8d0c8e5f29413 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -5080,28 +5080,45 @@ def shift(
         axis = self._get_axis_number(axis)
 
         ncols = len(self.columns)
-        if axis == 1 and periods != 0 and fill_value is lib.no_default and ncols > 0:
-            # We will infer fill_value to match the closest column
 
-            # Use a column that we know is valid for our column's dtype GH#38434
-            label = self.columns[0]
+        if (
+            axis == 1
+            and periods != 0
+            and ncols > 0
+            and (fill_value is lib.no_default or len(self._mgr.arrays) > 1)
+        ):
+            # Exclude single-array-with-fill_value case so we issue a FutureWarning
+            #  if an integer is passed with datetimelike dtype GH#31971
+            from pandas import concat
 
+            # tail: the data that is still in our shifted DataFrame
             if periods > 0:
-                result = self.iloc[:, :-periods]
-                for col in range(min(ncols, abs(periods))):
-                    # TODO(EA2D): doing this in a loop unnecessary with 2D EAs
-                    # Define filler inside loop so we get a copy
-                    filler = self.iloc[:, 0].shift(len(self))
-                    result.insert(0, label, filler, allow_duplicates=True)
+                tail = self.iloc[:, :-periods]
             else:
-                result = self.iloc[:, -periods:]
-                for col in range(min(ncols, abs(periods))):
-                    # Define filler inside loop so we get a copy
-                    filler = self.iloc[:, -1].shift(len(self))
-                    result.insert(
-                        len(result.columns), label, filler, allow_duplicates=True
-                    )
+                tail = self.iloc[:, -periods:]
+            # pin a simple Index to avoid costly casting
+            tail.columns = range(len(tail.columns))
+
+            if fill_value is not lib.no_default:
+                # GH#35488
+                # TODO(EA2D): with 2D EAs we could construct other directly
+                ser = Series(fill_value, index=self.index)
+            else:
+                # We infer fill_value to match the closest column
+                if periods > 0:
+                    ser = self.iloc[:, 0].shift(len(self))
+                else:
+                    ser = self.iloc[:, -1].shift(len(self))
+
+            width = min(abs(periods), ncols)
+            other = concat([ser] * width, axis=1)
+
+            if periods > 0:
+                result = concat([other, tail], axis=1)
+            else:
+                result = concat([tail, other], axis=1)
 
+            result = cast(DataFrame, result)
             result.columns = self.columns.copy()
             return result
 
diff --git a/pandas/core/internals/managers.py b/pandas/core/internals/managers.py
index 759500827f344..e4cce731b7b56 100644
--- a/pandas/core/internals/managers.py
+++ b/pandas/core/internals/managers.py
@@ -620,25 +620,6 @@ def shift(self, periods: int, axis: int, fill_value) -> BlockManager:
         if fill_value is lib.no_default:
             fill_value = None
 
-        if axis == 0 and self.ndim == 2 and self.nblocks > 1:
-            # GH#35488 we need to watch out for multi-block cases
-            # We only get here with fill_value not-lib.no_default
-            ncols = self.shape[0]
-            if periods > 0:
-                indexer = [-1] * periods + list(range(ncols - periods))
-            else:
-                nper = abs(periods)
-                indexer = list(range(nper, ncols)) + [-1] * nper
-            result = self.reindex_indexer(
-                self.items,
-                indexer,
-                axis=0,
-                fill_value=fill_value,
-                allow_dups=True,
-                consolidate=False,
-            )
-            return result
-
         return self.apply("shift", periods=periods, axis=axis, fill_value=fill_value)
 
     def fillna(self, value, limit, inplace: bool, downcast) -> BlockManager:
diff --git a/pandas/tests/apply/test_frame_transform.py b/pandas/tests/apply/test_frame_transform.py
index 7a626ce6312c5..6a3634ca8b62a 100644
--- a/pandas/tests/apply/test_frame_transform.py
+++ b/pandas/tests/apply/test_frame_transform.py
@@ -39,15 +39,8 @@ def test_transform_ufunc(axis, float_frame, frame_or_series):
 
 
 @pytest.mark.parametrize("op", frame_transform_kernels)
-def test_transform_groupby_kernel(axis, float_frame, op, using_array_manager, request):
+def test_transform_groupby_kernel(axis, float_frame, op, request):
     # GH 35964
-    if using_array_manager and op == "pct_change" and axis in (1, "columns"):
-        # TODO(ArrayManager) shift with axis=1
-        request.node.add_marker(
-            pytest.mark.xfail(
-                reason="shift axis=1 not yet implemented for ArrayManager"
-            )
-        )
 
     args = [0.0] if op == "fillna" else []
     if axis == 0 or axis == "index":