diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 638129291b495..672da40572eea 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -6614,11 +6614,11 @@ def _join_compat(self, other, on=None, how='left', lsuffix='', rsuffix='',
             if can_concat:
                 if how == 'left':
                     how = 'outer'
-                    join_axes = [self.index]
+                    return concat(frames, axis=1, join=how,
+                                  verify_integrity=True).reindex(self.index)
                 else:
-                    join_axes = None
-                return concat(frames, axis=1, join=how, join_axes=join_axes,
-                              verify_integrity=True)
+                    return concat(frames, axis=1, join=how,
+                                  verify_integrity=True)
 
             joined = frames[0]
 
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index ac7278953e658..138b2f09b1496 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -8848,7 +8848,7 @@ def describe_1d(data):
                 if name not in names:
                     names.append(name)
 
-        d = pd.concat(ldesc, join_axes=pd.Index([names]), axis=1)
+        d = pd.concat([x.reindex(names) for x in ldesc], axis=1, sort=False)
         d.columns = data.columns.copy()
         return d
 
diff --git a/pandas/core/groupby/generic.py b/pandas/core/groupby/generic.py
index 685635fb6854d..7c64c8d9424ca 100644
--- a/pandas/core/groupby/generic.py
+++ b/pandas/core/groupby/generic.py
@@ -518,8 +518,10 @@ def _transform_general(self, func, *args, **kwargs):
                 applied.append(res)
 
         concat_index = obj.columns if self.axis == 0 else obj.index
-        concatenated = concat(applied, join_axes=[concat_index],
-                              axis=self.axis, verify_integrity=False)
+        other_axis = (self.axis + 1) % 2  # switches from 0 to 1 or from 1 to 0
+        concatenated = concat(applied, axis=self.axis,
+                              verify_integrity=False).reindex(concat_index,
+                                                              axis=other_axis)
         return self._set_result_index_ordered(concatenated)
 
     @Substitution(klass='DataFrame', selected='')