diff --git a/pkg/controller/podautoscaler/horizontal.go b/pkg/controller/podautoscaler/horizontal.go
index 3b179a89077d0..81dcc17245cbc 100644
--- a/pkg/controller/podautoscaler/horizontal.go
+++ b/pkg/controller/podautoscaler/horizontal.go
@@ -553,6 +553,7 @@ func (a *HorizontalController) computeStatusForObjectMetric(specReplicas, status
 		*status = autoscalingv2.MetricStatus{
 			Type: autoscalingv2.ObjectMetricSourceType,
 			Object: &autoscalingv2.ObjectMetricStatus{
+				DescribedObject: metricSpec.Object.DescribedObject,
 				Metric: autoscalingv2.MetricIdentifier{
 					Name:     metricSpec.Object.Metric.Name,
 					Selector: metricSpec.Object.Metric.Selector,
diff --git a/pkg/controller/podautoscaler/horizontal_test.go b/pkg/controller/podautoscaler/horizontal_test.go
index 0f2fb2ed99d9d..79f17ddb5adf9 100644
--- a/pkg/controller/podautoscaler/horizontal_test.go
+++ b/pkg/controller/podautoscaler/horizontal_test.go
@@ -381,6 +381,13 @@ func (tc *testCase) prepareTestClient(t *testing.T) (*fake.Clientset, *metricsfa
 					assert.Equal(t, tc.CPUCurrent, *utilization, "the report CPU utilization percentage should be as expected")
 				}
 			}
+
+			if len(obj.Spec.Metrics) > 0 && obj.Spec.Metrics[0].Object != nil && len(obj.Status.CurrentMetrics) > 0 && obj.Status.CurrentMetrics[0].Object != nil {
+				assert.Equal(t, obj.Spec.Metrics[0].Object.DescribedObject.APIVersion, obj.Status.CurrentMetrics[0].Object.DescribedObject.APIVersion)
+				assert.Equal(t, obj.Spec.Metrics[0].Object.DescribedObject.Kind, obj.Status.CurrentMetrics[0].Object.DescribedObject.Kind)
+				assert.Equal(t, obj.Spec.Metrics[0].Object.DescribedObject.Name, obj.Status.CurrentMetrics[0].Object.DescribedObject.Name)
+			}
+
 			actualConditions := obj.Status.Conditions
 			// TODO: it's ok not to sort these because statusOk
 			// contains all the conditions, so we'll never be appending.