diff --git a/doc/source/whatsnew/v1.3.0.rst b/doc/source/whatsnew/v1.3.0.rst
index 99ae60859b68c..11f8a5b6c2499 100644
--- a/doc/source/whatsnew/v1.3.0.rst
+++ b/doc/source/whatsnew/v1.3.0.rst
@@ -390,6 +390,7 @@ I/O
 - Bug in :func:`read_json` when ``orient="split"`` does not maintain numeric string index (:issue:`28556`)
 - :meth:`read_sql` returned an empty generator if ``chunksize`` was no-zero and the query returned no results. Now returns a generator with a single empty dataframe (:issue:`34411`)
 - Bug in :func:`read_hdf` returning unexpected records when filtering on categorical string columns using ``where`` parameter (:issue:`39189`)
+- Bug in :func:`read_sas` raising ``ValueError`` when ``datetimes`` were null (:issue:`39725`)
 
 Period
 ^^^^^^
diff --git a/pandas/io/sas/sas7bdat.py b/pandas/io/sas/sas7bdat.py
index fe06f103e6c5e..9853fa41d3fb9 100644
--- a/pandas/io/sas/sas7bdat.py
+++ b/pandas/io/sas/sas7bdat.py
@@ -23,6 +23,7 @@
 from pandas.errors import EmptyDataError, OutOfBoundsDatetime
 
 import pandas as pd
+from pandas import isna
 
 from pandas.io.common import get_handle
 from pandas.io.sas._sas import Parser
@@ -30,6 +31,20 @@
 from pandas.io.sas.sasreader import ReaderBase
 
 
+def _parse_datetime(sas_datetime: float, unit: str):
+    if isna(sas_datetime):
+        return pd.NaT
+
+    if unit == "s":
+        return datetime(1960, 1, 1) + timedelta(seconds=sas_datetime)
+
+    elif unit == "d":
+        return datetime(1960, 1, 1) + timedelta(days=sas_datetime)
+
+    else:
+        raise ValueError("unit must be 'd' or 's'")
+
+
 def _convert_datetimes(sas_datetimes: pd.Series, unit: str) -> pd.Series:
     """
     Convert to Timestamp if possible, otherwise to datetime.datetime.
@@ -51,20 +66,9 @@ def _convert_datetimes(sas_datetimes: pd.Series, unit: str) -> pd.Series:
     try:
         return pd.to_datetime(sas_datetimes, unit=unit, origin="1960-01-01")
     except OutOfBoundsDatetime:
-        if unit == "s":
-            s_series = sas_datetimes.apply(
-                lambda sas_float: datetime(1960, 1, 1) + timedelta(seconds=sas_float)
-            )
-            s_series = cast(pd.Series, s_series)
-            return s_series
-        elif unit == "d":
-            d_series = sas_datetimes.apply(
-                lambda sas_float: datetime(1960, 1, 1) + timedelta(days=sas_float)
-            )
-            d_series = cast(pd.Series, d_series)
-            return d_series
-        else:
-            raise ValueError("unit must be 'd' or 's'")
+        s_series = sas_datetimes.apply(_parse_datetime, unit=unit)
+        s_series = cast(pd.Series, s_series)
+        return s_series
 
 
 class _SubheaderPointer:
diff --git a/pandas/tests/io/sas/data/dates_null.sas7bdat b/pandas/tests/io/sas/data/dates_null.sas7bdat
new file mode 100644
index 0000000000000..beadf1a34f42e
Binary files /dev/null and b/pandas/tests/io/sas/data/dates_null.sas7bdat differ
diff --git a/pandas/tests/io/sas/test_sas7bdat.py b/pandas/tests/io/sas/test_sas7bdat.py
index 4035f6aff90ce..b23959a7d87a2 100644
--- a/pandas/tests/io/sas/test_sas7bdat.py
+++ b/pandas/tests/io/sas/test_sas7bdat.py
@@ -315,3 +315,22 @@ def test_max_sas_date_iterator(datapath):
     ]
     for result, expected in zip(results, expected):
         tm.assert_frame_equal(result, expected)
+
+
+def test_null_date(datapath):
+    fname = datapath("io", "sas", "data", "dates_null.sas7bdat")
+    df = pd.read_sas(fname, encoding="utf-8")
+
+    expected = pd.DataFrame(
+        {
+            "datecol": [
+                datetime(9999, 12, 29),
+                pd.NaT,
+            ],
+            "datetimecol": [
+                datetime(9999, 12, 29, 23, 59, 59, 998993),
+                pd.NaT,
+            ],
+        },
+    )
+    tm.assert_frame_equal(df, expected)