diff --git a/doc/source/whatsnew/v2.0.0.rst b/doc/source/whatsnew/v2.0.0.rst
index 13b60f7f05352..59a27b0969a1e 100644
--- a/doc/source/whatsnew/v2.0.0.rst
+++ b/doc/source/whatsnew/v2.0.0.rst
@@ -951,6 +951,7 @@ Groupby/resample/rolling
 - Bug in :meth:`.SeriesGroupBy.nth` would raise when grouper contained NA values after subsetting from a :class:`DataFrameGroupBy` (:issue:`26454`)
 - Bug in :meth:`DataFrame.groupby` would not include a :class:`.Grouper` specified by ``key`` in the result when ``as_index=False`` (:issue:`50413`)
 - Bug in :meth:`.DataFrameGrouBy.value_counts` would raise when used with a :class:`.TimeGrouper` (:issue:`50486`)
+- Bug in :meth:`Resampler.size` caused a wide :class:`DataFrame` to be returned instead of a :class:`Series` with :class:`MultiIndex` (:issue:`46826`)
 -
 
 Reshaping
diff --git a/pandas/core/resample.py b/pandas/core/resample.py
index 4af4be20a056e..d6cba824767b5 100644
--- a/pandas/core/resample.py
+++ b/pandas/core/resample.py
@@ -989,6 +989,12 @@ def var(
     @doc(GroupBy.size)
     def size(self):
         result = self._downsample("size")
+
+        # If the result is a non-empty DataFrame we stack to get a Series
+        # GH 46826
+        if isinstance(result, ABCDataFrame) and not result.empty:
+            result = result.stack()
+
         if not len(self.ax):
             from pandas import Series
 
diff --git a/pandas/tests/resample/test_resampler_grouper.py b/pandas/tests/resample/test_resampler_grouper.py
index 0432cf397067d..a521e24aa6022 100644
--- a/pandas/tests/resample/test_resampler_grouper.py
+++ b/pandas/tests/resample/test_resampler_grouper.py
@@ -515,3 +515,25 @@ def test_resample_empty_Dataframe(keys):
         expected.index.name = keys[0]
 
     tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_resample_size_all_index_same():
+    # GH 46826
+    df = DataFrame(
+        {"A": [1] * 3 + [2] * 3 + [1] * 3 + [2] * 3, "B": np.arange(12)},
+        index=date_range("31/12/2000 18:00", freq="H", periods=12),
+    )
+    result = df.groupby("A").resample("D").size()
+    expected = Series(
+        3,
+        index=pd.MultiIndex.from_tuples(
+            [
+                (1, Timestamp("2000-12-31")),
+                (1, Timestamp("2001-01-01")),
+                (2, Timestamp("2000-12-31")),
+                (2, Timestamp("2001-01-01")),
+            ],
+            names=["A", None],
+        ),
+    )
+    tm.assert_series_equal(result, expected)