diff --git a/.pre-commit-hooks.yaml b/.pre-commit-hooks.yaml
index 4b4d0cf..456a4f2 100644
--- a/.pre-commit-hooks.yaml
+++ b/.pre-commit-hooks.yaml
@@ -144,12 +144,6 @@
     entry: fix-byte-order-marker
     language: python
     types: [text]
--   id: fix-encoding-pragma
-    name: fix python encoding pragma (deprecated)
-    description: 'adds # -*- coding: utf-8 -*- to the top of python files.'
-    language: python
-    entry: fix-encoding-pragma
-    types: [python]
 -   id: forbid-new-submodules
     name: forbid new submodules
     description: prevents addition of new git submodules.
diff --git a/README.md b/README.md
index 4992baf..9bcda1b 100644
--- a/README.md
+++ b/README.md
@@ -126,13 +126,6 @@ The following arguments are available:
 #### `fix-byte-order-marker`
 removes UTF-8 byte order marker
 
-#### `fix-encoding-pragma`
-
-_Deprecated since py2 is EOL - use [pyupgrade](https://github.com/asottile/pyupgrade) instead._
-
-Add `# -*- coding: utf-8 -*-` to the top of python files.
-  - To remove the coding pragma pass `--remove` (useful in a python3-only codebase)
-
 #### `forbid-new-submodules`
 Prevent addition of new git submodules.
 
@@ -210,6 +203,7 @@ Trims trailing whitespace.
 ### Deprecated / replaced hooks
 
 - `check-byte-order-marker`: instead use fix-byte-order-marker
+- `fix-encoding-pragma`: instead use [`pyupgrade`](https://github.com/asottile/pyupgrade)
 
 ### As a standalone package
 
diff --git a/pre_commit_hooks/fix_encoding_pragma.py b/pre_commit_hooks/fix_encoding_pragma.py
deleted file mode 100644
index eee6705..0000000
--- a/pre_commit_hooks/fix_encoding_pragma.py
+++ /dev/null
@@ -1,157 +0,0 @@
-from __future__ import annotations
-
-import argparse
-import sys
-from typing import IO
-from typing import NamedTuple
-from typing import Sequence
-
-DEFAULT_PRAGMA = b'# -*- coding: utf-8 -*-'
-
-
-def has_coding(line: bytes) -> bool:
-    if not line.strip():
-        return False
-    return (
-        line.lstrip()[:1] == b'#' and (
-            b'unicode' in line or
-            b'encoding' in line or
-            b'coding:' in line or
-            b'coding=' in line
-        )
-    )
-
-
-class ExpectedContents(NamedTuple):
-    shebang: bytes
-    rest: bytes
-    # True: has exactly the coding pragma expected
-    # False: missing coding pragma entirely
-    # None: has a coding pragma, but it does not match
-    pragma_status: bool | None
-    ending: bytes
-
-    @property
-    def has_any_pragma(self) -> bool:
-        return self.pragma_status is not False
-
-    def is_expected_pragma(self, remove: bool) -> bool:
-        expected_pragma_status = not remove
-        return self.pragma_status is expected_pragma_status
-
-
-def _get_expected_contents(
-        first_line: bytes,
-        second_line: bytes,
-        rest: bytes,
-        expected_pragma: bytes,
-) -> ExpectedContents:
-    ending = b'\r\n' if first_line.endswith(b'\r\n') else b'\n'
-
-    if first_line.startswith(b'#!'):
-        shebang = first_line
-        potential_coding = second_line
-    else:
-        shebang = b''
-        potential_coding = first_line
-        rest = second_line + rest
-
-    if potential_coding.rstrip(b'\r\n') == expected_pragma:
-        pragma_status: bool | None = True
-    elif has_coding(potential_coding):
-        pragma_status = None
-    else:
-        pragma_status = False
-        rest = potential_coding + rest
-
-    return ExpectedContents(
-        shebang=shebang, rest=rest, pragma_status=pragma_status, ending=ending,
-    )
-
-
-def fix_encoding_pragma(
-        f: IO[bytes],
-        remove: bool = False,
-        expected_pragma: bytes = DEFAULT_PRAGMA,
-) -> int:
-    expected = _get_expected_contents(
-        f.readline(), f.readline(), f.read(), expected_pragma,
-    )
-
-    # Special cases for empty files
-    if not expected.rest.strip():
-        # If a file only has a shebang or a coding pragma, remove it
-        if expected.has_any_pragma or expected.shebang:
-            f.seek(0)
-            f.truncate()
-            f.write(b'')
-            return 1
-        else:
-            return 0
-
-    if expected.is_expected_pragma(remove):
-        return 0
-
-    # Otherwise, write out the new file
-    f.seek(0)
-    f.truncate()
-    f.write(expected.shebang)
-    if not remove:
-        f.write(expected_pragma + expected.ending)
-    f.write(expected.rest)
-
-    return 1
-
-
-def _normalize_pragma(pragma: str) -> bytes:
-    return pragma.encode().rstrip()
-
-
-def main(argv: Sequence[str] | None = None) -> int:
-    print(
-        'warning: this hook is deprecated and will be removed in a future '
-        'release because py2 is EOL. instead, use '
-        'https://github.com/asottile/pyupgrade',
-        file=sys.stderr,
-    )
-
-    parser = argparse.ArgumentParser(
-        'Fixes the encoding pragma of python files',
-    )
-    parser.add_argument('filenames', nargs='*', help='Filenames to fix')
-    parser.add_argument(
-        '--pragma', default=DEFAULT_PRAGMA, type=_normalize_pragma,
-        help=(
-            f'The encoding pragma to use.  '
-            f'Default: {DEFAULT_PRAGMA.decode()}'
-        ),
-    )
-    parser.add_argument(
-        '--remove', action='store_true',
-        help='Remove the encoding pragma (Useful in a python3-only codebase)',
-    )
-    args = parser.parse_args(argv)
-
-    retv = 0
-
-    if args.remove:
-        fmt = 'Removed encoding pragma from {filename}'
-    else:
-        fmt = 'Added `{pragma}` to {filename}'
-
-    for filename in args.filenames:
-        with open(filename, 'r+b') as f:
-            file_ret = fix_encoding_pragma(
-                f, remove=args.remove, expected_pragma=args.pragma,
-            )
-            retv |= file_ret
-            if file_ret:
-                print(
-                    fmt.format(pragma=args.pragma.decode(), filename=filename),
-                )
-
-    return retv
-
-
-if __name__ == '__main__':
-    raise SystemExit(main())
diff --git a/tests/fix_encoding_pragma_test.py b/tests/fix_encoding_pragma_test.py
deleted file mode 100644
index 98557e9..0000000
--- a/tests/fix_encoding_pragma_test.py
+++ /dev/null
@@ -1,161 +0,0 @@
-from __future__ import annotations
-
-import io
-
-import pytest
-
-from pre_commit_hooks.fix_encoding_pragma import _normalize_pragma
-from pre_commit_hooks.fix_encoding_pragma import fix_encoding_pragma
-from pre_commit_hooks.fix_encoding_pragma import main
-
-
-def test_integration_inserting_pragma(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'import httplib\n')
-
-    assert main((str(path),)) == 1
-
-    assert path.read_binary() == (
-        b'# -*- coding: utf-8 -*-\n'
-        b'import httplib\n'
-    )
-
-
-def test_integration_ok(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'# -*- coding: utf-8 -*-\nx = 1\n')
-    assert main((str(path),)) == 0
-
-
-def test_integration_remove(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'# -*- coding: utf-8 -*-\nx = 1\n')
-
-    assert main((str(path), '--remove')) == 1
-
-    assert path.read_binary() == b'x = 1\n'
-
-
-def test_integration_remove_ok(tmpdir):
-    path = tmpdir.join('foo.py')
-    path.write_binary(b'x = 1\n')
-    assert main((str(path), '--remove')) == 0
-
-
-@pytest.mark.parametrize(
-    'input_str',
-    (
-        b'',
-        (
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n'
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'foo = "bar"\n'
-        ),
-    ),
-)
-def test_ok_inputs(input_str):
-    bytesio = io.BytesIO(input_str)
-    assert fix_encoding_pragma(bytesio) == 0
-    bytesio.seek(0)
-    assert bytesio.read() == input_str
-
-
-@pytest.mark.parametrize(
-    ('input_str', 'output'),
-    (
-        (
-            b'import httplib\n',
-            b'# -*- coding: utf-8 -*-\n'
-            b'import httplib\n',
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'x = 1\n',
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        (
-            b'#coding=utf-8\n'
-            b'x = 1\n',
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        (
-            b'#!/usr/bin/env python\n'
-            b'#coding=utf8\n'
-            b'x = 1\n',
-            b'#!/usr/bin/env python\n'
-            b'# -*- coding: utf-8 -*-\n'
-            b'x = 1\n',
-        ),
-        # These should each get truncated
-        (b'#coding: utf-8\n', b''),
-        (b'# -*- coding: utf-8 -*-\n', b''),
-        (b'#!/usr/bin/env python\n', b''),
-        (b'#!/usr/bin/env python\n#coding: utf8\n', b''),
-        (b'#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n', b''),
-    ),
-)
-def test_not_ok_inputs(input_str, output):
-    bytesio = io.BytesIO(input_str)
-    assert fix_encoding_pragma(bytesio) == 1
-    bytesio.seek(0)
-    assert bytesio.read() == output
-
-
-def test_ok_input_alternate_pragma():
-    input_s = b'# coding: utf-8\nx = 1\n'
-    bytesio = io.BytesIO(input_s)
-    ret = fix_encoding_pragma(bytesio, expected_pragma=b'# coding: utf-8')
-    assert ret == 0
-    bytesio.seek(0)
-    assert bytesio.read() == input_s
-
-
-def test_not_ok_input_alternate_pragma():
-    bytesio = io.BytesIO(b'x = 1\n')
-    ret = fix_encoding_pragma(bytesio, expected_pragma=b'# coding: utf-8')
-    assert ret == 1
-    bytesio.seek(0)
-    assert bytesio.read() == b'# coding: utf-8\nx = 1\n'
-
-
-@pytest.mark.parametrize(
-    ('input_s', 'expected'),
-    (
-        ('# coding: utf-8', b'# coding: utf-8'),
-        # trailing whitespace
-        ('# coding: utf-8\n', b'# coding: utf-8'),
-    ),
-)
-def test_normalize_pragma(input_s, expected):
-    assert _normalize_pragma(input_s) == expected
-
-
-def test_integration_alternate_pragma(tmpdir, capsys):
-    f = tmpdir.join('f.py')
-    f.write('x = 1\n')
-
-    pragma = '# coding: utf-8'
-    assert main((str(f), '--pragma', pragma)) == 1
-    assert f.read() == '# coding: utf-8\nx = 1\n'
-    out, _ = capsys.readouterr()
-    assert out == f'Added `# coding: utf-8` to {str(f)}\n'
-
-
-def test_crlf_ok(tmpdir):
-    f = tmpdir.join('f.py')
-    f.write_binary(b'# -*- coding: utf-8 -*-\r\nx = 1\r\n')
-    assert not main((str(f),))
-
-
-def test_crfl_adds(tmpdir):
-    f = tmpdir.join('f.py')
-    f.write_binary(b'x = 1\r\n')
-    assert main((str(f),))
-    assert f.read_binary() == b'# -*- coding: utf-8 -*-\r\nx = 1\r\n'