diff --git a/pavelib/paver_tests/test_paver_quality.py b/pavelib/paver_tests/test_paver_quality.py
index 8024ca8255..6ae2942b99 100644
--- a/pavelib/paver_tests/test_paver_quality.py
+++ b/pavelib/paver_tests/test_paver_quality.py
@@ -16,6 +16,7 @@ from paver.easy import BuildFailure  # pylint: disable=ungrouped-imports
 
 import pavelib.quality
 from pavelib.paver_tests.utils import fail_on_eslint
+from pavelib.paver_tests.utils import PaverTestCase
 
 
 @ddt
@@ -268,7 +269,7 @@ class TestPrepareReportDir(unittest.TestCase):
         self.assertEqual(os.listdir(path(self.test_dir)), [])
 
 
-class TestPaverRunQuality(unittest.TestCase):
+class TestPaverRunQuality(PaverTestCase):
     """
     For testing the paver run_quality task
     """
@@ -276,19 +277,6 @@ class TestPaverRunQuality(unittest.TestCase):
     def setUp(self):
         super(TestPaverRunQuality, self).setUp()
 
-        # test_no_diff_quality_failures seems to alter the way that paver
-        # executes these lines is subsequent tests.
-        # https://github.com/paver/paver/blob/master/paver/tasks.py#L175-L180
-        #
-        # The other tests don't appear to have the same impact. This was
-        # causing a test order dependency. This line resets that state
-        # of environment._task_in_progress so that the paver commands in the
-        # tests will be considered top level tasks by paver, and we can predict
-        # which path it will chose in the above code block.
-        #
-        # TODO: Figure out why one test is altering the state to begin with.
-        paver.tasks.environment = paver.tasks.Environment()
-
         # mock the @needs decorator to skip it
         patcher = patch('pavelib.quality.sh')
         self._mock_paver_sh = patcher.start()
@@ -326,7 +314,6 @@ class TestPaverRunQuality(unittest.TestCase):
         """
         If diff-quality fails on eslint, the paver task should also fail
         """
-
         # Underlying sh call must fail when it is running the eslint diff-quality task
         self._mock_paver_sh.side_effect = fail_on_eslint
         _mock_pylint_violations = MagicMock(return_value=(0, []))
diff --git a/pavelib/paver_tests/utils.py b/pavelib/paver_tests/utils.py
index 3fb9ea55aa..ab0f1b9d3d 100644
--- a/pavelib/paver_tests/utils.py
+++ b/pavelib/paver_tests/utils.py
@@ -70,21 +70,7 @@ def fail_on_eslint(*args):
     percentage ("p") requirement.
     """
     if "eslint" in args[0]:
-        # Essentially mock diff-quality exiting with 1
-        paver.easy.sh("exit 1")
-    else:
-        return
-
-
-def fail_on_pylint(*args):
-    """
-    For our tests, we need the call for diff-quality running pylint reports
-    to fail, since that is what is going to fail when we pass in a
-    percentage ("p") requirement.
-    """
-    if "pylint" in args[0]:
-        # Essentially mock diff-quality exiting with 1
-        paver.easy.sh("exit 1")
+        raise BuildFailure('Subprocess return code: 1')
     else:
         return