diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
index 8cf407382619..4db5c48ef3bc 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
@@ -486,14 +486,16 @@ class Analyzer(
       case Pivot(groupByExprs, pivotColumn, pivotValues, aggregates, child) =>
         val singleAgg = aggregates.size == 1
         def outputName(value: Literal, aggregate: Expression): String = {
+          val utf8Value = Cast(value, StringType, Some(conf.sessionLocalTimeZone)).eval(EmptyRow)
+          val stringValue: String = Option(utf8Value).map(_.toString).getOrElse("null")
           if (singleAgg) {
-            value.toString
+            stringValue
           } else {
             val suffix = aggregate match {
               case n: NamedExpression => n.name
               case _ => toPrettySQL(aggregate)
             }
-            value + "_" + suffix
+            stringValue + "_" + suffix
           }
         }
         if (aggregates.forall(a => PivotFirst.supportsDataType(a.dataType))) {
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
index ca3cb5676742..6ca9ee57e8f4 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
@@ -23,7 +23,7 @@ import org.apache.spark.sql.internal.SQLConf
 import org.apache.spark.sql.test.SharedSQLContext
 import org.apache.spark.sql.types._
 
-class DataFramePivotSuite extends QueryTest with SharedSQLContext{
+class DataFramePivotSuite extends QueryTest with SharedSQLContext {
   import testImplicits._
 
   test("pivot courses") {
@@ -230,4 +230,20 @@ class DataFramePivotSuite extends QueryTest with SharedSQLContext{
         .groupBy($"a").pivot("a").agg(min($"b")),
       Row(null, Seq(null, 7), null) :: Row(1, null, Seq(1, 7)) :: Nil)
   }
+
+  test("pivot with timestamp and count should not print internal representation") {
+    val ts = "2012-12-31 16:00:10.011"
+    val tsWithZone = "2013-01-01 00:00:10.011"
+
+    withSQLConf(SQLConf.SESSION_LOCAL_TIMEZONE.key -> "GMT") {
+      val df = Seq(java.sql.Timestamp.valueOf(ts)).toDF("a").groupBy("a").pivot("a").count()
+      val expected = StructType(
+        StructField("a", TimestampType) ::
+        StructField(tsWithZone, LongType) :: Nil)
+      assert(df.schema == expected)
+      // String representation of timestamp with timezone should take the time difference
+      // into account.
+      checkAnswer(df.select($"a".cast(StringType)), Row(tsWithZone))
+    }
+  }
 }