diff --git a/pdfannots/cli.py b/pdfannots/cli.py
index 9e926fe..58692af 100644
--- a/pdfannots/cli.py
+++ b/pdfannots/cli.py
@@ -14,6 +14,7 @@
 MD_FORMAT_ARGS = frozenset([
     'condense',
     'group_highlights_by_color',
+    'page_number_offset',
     'print_filename',
     'sections',
     'use_page_labels',
@@ -77,6 +78,8 @@ def parse_args() -> typ.Tuple[argparse.Namespace, LAParams]:
                    help="Emit annotations as a blockquote regardless of length.")
     g.add_argument("--no-page-labels", dest="use_page_labels", default=True, action="store_false",
                    help="Ignore page labels if present, just print 1-based page numbers.")
+    g.add_argument("--page-number-offset", dest="page_number_offset", default=0, type=int,
+                   help="Increase or decrease page numbers with a fixed offset.")
     g.add_argument("--print-filename", dest="print_filename", default=False, action="store_true",
                    help="Print the name of each file with annotations.")
     g.add_argument("-w", "--wrap", dest="wrap_column", metavar="COLS", type=int,
diff --git a/pdfannots/printer/markdown.py b/pdfannots/printer/markdown.py
index cf790dc..27c8889 100644
--- a/pdfannots/printer/markdown.py
+++ b/pdfannots/printer/markdown.py
@@ -86,12 +86,14 @@ def __init__(
         self,
         *,
         condense: bool = True,                  # Permit use of the condensed format
+        page_number_offset: int = 0,            # Page number offset
         print_filename: bool = False,           # Whether to print file names
         remove_hyphens: bool = True,            # Whether to remove hyphens across a line break
         use_page_labels: bool = True,           # Whether to use page labels
         wrap_column: typ.Optional[int] = None,  # Column at which output is word-wrapped
         **kwargs: typ.Any                       # Other args, ignored
     ) -> None:
+        self.page_number_offset = page_number_offset
         self.print_filename = print_filename
         self.remove_hyphens = remove_hyphens
         self.use_page_labels = use_page_labels
@@ -140,10 +142,13 @@ def print_file(
     def format_pos(
         pos: Pos,
         document: Document,
-        use_page_label: bool
+        use_page_label: bool,
+        page_number_offset: int
     ) -> str:
 
-        result = pos.page.format_name(use_label=use_page_label).title()
+        result = pos.page.format_name(
+            use_label=use_page_label,
+            page_number_offset=page_number_offset).title()
 
         o = document.nearest_outline(pos)
         if o:
@@ -230,8 +235,9 @@ def format_annot(
 
         # compute the formatted position (and extra bit if needed) as a label
         assert annot.pos is not None
-        label = self.format_pos(annot.pos, document, self.use_page_labels) + \
-            (" " + extra if extra else "") + ":"
+        label = self.format_pos(
+            annot.pos, document, self.use_page_labels, self.page_number_offset
+        ) + (" " + extra if extra else "") + ":"
 
         # If we have short (few words) text with a short or no comment, and the
         # text contains no embedded full stops or quotes, then we'll just put
diff --git a/pdfannots/types.py b/pdfannots/types.py
index a7b7531..b61fba4 100644
--- a/pdfannots/types.py
+++ b/pdfannots/types.py
@@ -133,12 +133,12 @@ def __repr__(self) -> str:
     def __str__(self) -> str:
         return self.format_name()
 
-    def format_name(self, use_label: bool = True) -> str:
+    def format_name(self, use_label: bool = True, page_number_offset: int = 0) -> str:
         if self.label and use_label:
             return 'page %s' % self.label
         else:
             # + 1 for 1-based page numbers in normal program output (error messages, etc.)
-            return 'page #%d' % (self.pageno + 1)
+            return 'page #%d' % (self.pageno + 1 + page_number_offset)
 
     def __eq__(self, other: object) -> bool:
         if not isinstance(other, Page):
diff --git a/tests.py b/tests.py
index 174b0df..9fb3363 100755
--- a/tests.py
+++ b/tests.py
@@ -4,6 +4,7 @@
 import json
 import operator
 import pathlib
+import re
 import typing as typ
 import unittest
 from datetime import datetime, timedelta, timezone
@@ -280,6 +281,17 @@ def test_flat(self) -> None:
         self.assertGreater(linecount, 5)
         self.assertGreater(charcount, 500)
 
+    def test_flat_page_number_offset(self) -> None:
+        p = MarkdownPrinter(page_number_offset=-1)
+
+        page_numbers = []
+        for line in p.print_file('dummyfile', self.doc):
+            m = re.match(r'.+Page #([0-9])', line)
+            if m:
+                page_numbers.append(m[1])
+
+        self.assertEqual(page_numbers, ['0', '0', '1', '1', '1', '1', '3', '3', '3'])
+
     def test_grouped(self) -> None:
         p = GroupedMarkdownPrinter(wrap_column=80)