diff --git a/quantecon/game_theory/__init__.py b/quantecon/game_theory/__init__.py
index 03249c3c..7ca07de5 100644
--- a/quantecon/game_theory/__init__.py
+++ b/quantecon/game_theory/__init__.py
@@ -24,3 +24,4 @@
 from .logitdyn import LogitDynamics
 from .polymatrix_game import PolymatrixGame
 from .howson_lcp import polym_lcp_solver
+from .game_converters import GAMReader, GAMWriter, from_gam, to_gam
diff --git a/quantecon/game_theory/game_converters.py b/quantecon/game_theory/game_converters.py
index 8218a3d6..c5b305a5 100644
--- a/quantecon/game_theory/game_converters.py
+++ b/quantecon/game_theory/game_converters.py
@@ -7,8 +7,11 @@
 Create a QuantEcon NormalFormGame from a gam file storing
 a 3-player Minimum Effort Game
 
->>> filepath = "./tests/gam_files/minimum_effort_game.gam"
->>> nfg = qe_nfg_from_gam_file(filepath)
+>>> import os
+>>> import quantecon.game_theory as gt
+>>> filepath = os.path.dirname(gt.__file__)
+>>> filepath += "/tests/game_files/minimum_effort_game.gam"
+>>> nfg = gt.from_gam(filepath)
 >>> print(nfg)
 3-player NormalFormGame with payoff profile array:
 [[[[  1.,   1.,   1.],   [  1.,   1.,  -9.],   [  1.,   1., -19.]],
@@ -24,12 +27,183 @@
   [[-19., -19.,   1.],   [ -8.,  -8.,   2.],   [  3.,   3.,   3.]]]]
 
 """
+import numpy as np
+from .normal_form_game import Player, NormalFormGame
 
-from .normal_form_game import NormalFormGame
-from itertools import product
 
+def str2num(s):
+    if '.' in s:
+        return float(s)
+    return int(s)
 
-def qe_nfg_from_gam_file(filename: str) -> NormalFormGame:
+
+class GAMReader:
+    """
+    Reader object that converts a game in GameTracer gam format into
+    a NormalFormGame.
+
+    """
+    @classmethod
+    def from_file(cls, file_path):
+        """
+        Read from a gam format file.
+
+        Parameters
+        ----------
+        file_path : str
+            Path to gam file.
+
+        Returns
+        -------
+        NormalFormGame
+
+        Examples
+        --------
+        Save a gam format string in a temporary file:
+
+        >>> import tempfile
+        >>> fname = tempfile.mkstemp()[1]
+        >>> with open(fname, mode='w') as f:
+        ...       f.write(\"\"\"\\
+        ... 2
+        ... 3 2
+        ...
+        ... 3 2 0 3 5 6 3 2 3 2 6 1\"\"\")
+
+        Read the file:
+
+        >>> g = GAMReader.from_file(fname)
+        >>> print(g)
+        2-player NormalFormGame with payoff profile array:
+        [[[3, 3],  [3, 2]],
+         [[2, 2],  [5, 6]],
+         [[0, 3],  [6, 1]]]
+
+        """
+        with open(file_path, 'r') as f:
+            string = f.read()
+        return cls._parse(string)
+
+    @classmethod
+    def from_url(cls, url):
+        """
+        Read from a URL.
+
+        """
+        import urllib.request
+        with urllib.request.urlopen(url) as response:
+            string = response.read().decode()
+        return cls._parse(string)
+
+    @classmethod
+    def from_string(cls, string):
+        """
+        Read from a gam format string.
+
+        Parameters
+        ----------
+        string : str
+            String in gam format.
+
+        Returns
+        -------
+        NormalFormGame
+
+        Examples
+        --------
+        >>> string = \"\"\"\\
+        ... 2
+        ... 3 2
+        ...
+        ... 3 2 0 3 5 6 3 2 3 2 6 1\"\"\"
+        >>> g = GAMReader.from_string(string)
+        >>> print(g)
+        2-player NormalFormGame with payoff profile array:
+        [[[3, 3],  [3, 2]],
+         [[2, 2],  [5, 6]],
+         [[0, 3],  [6, 1]]]
+
+        """
+        return cls._parse(string)
+
+    @staticmethod
+    def _parse(string):
+        tokens = string.split()
+
+        N = int(tokens.pop(0))
+        nums_actions = tuple(int(tokens.pop(0)) for _ in range(N))
+        payoffs = np.array([str2num(s) for s in tokens])
+
+        na = np.prod(nums_actions)
+        payoffs2d = payoffs.reshape(N, na)
+        players = [
+            Player(
+                payoffs2d[i, :].reshape(nums_actions, order='F').transpose(
+                    list(range(i, N)) + list(range(i))
+                )
+            ) for i in range(N)
+        ]
+
+        return NormalFormGame(players)
+
+
+class GAMWriter:
+    """
+    Writer object that converts a NormalFormgame into a game in
+    GameTracer gam format.
+
+    """
+    @classmethod
+    def to_file(cls, g, file_path):
+        """
+        Save the GameTracer gam format string representation of the
+        NormalFormGame `g` to a file.
+
+        Parameters
+        ----------
+        g : NormalFormGame
+
+        file_path : str
+            Path to the file to write to.
+
+        """
+        with open(file_path, 'w') as f:
+            f.write(cls._dump(g) + '\n')
+
+    @classmethod
+    def to_string(cls, g):
+        """
+        Return a GameTracer gam format string representing the
+        NormalFormGame `g`.
+
+        Parameters
+        ----------
+        g : NormalFormGame
+
+        Returns
+        -------
+        str
+            String representation in gam format.
+
+        """
+        return cls._dump(g)
+
+    @staticmethod
+    def _dump(g):
+        s = str(g.N) + '\n'
+        s += ' '.join(map(str, g.nums_actions)) + '\n\n'
+
+        for i, player in enumerate(g.players):
+            payoffs = np.array2string(
+                player.payoff_array.transpose(
+                    list(range(g.N-i, g.N)) + list(range(g.N-i))
+                ).ravel(order='F'))[1:-1]
+            s += ' '.join(payoffs.split()) + ' '
+
+        return s.rstrip()
+
+
+def from_gam(filename: str) -> NormalFormGame:
     """
     Makes a QuantEcon Normal Form Game from a gam file.
 
@@ -51,32 +225,22 @@ def qe_nfg_from_gam_file(filename: str) -> NormalFormGame:
        http://dags.stanford.edu/Games/gametracer.html
 
     """
-    with open(filename, 'r') as file:
-        lines = file.readlines()
-        combined = [
-            token
-            for line in lines
-            for token in line.split()
-        ]
+    return GAMReader.from_file(filename)
 
-        i = iter(combined)
-        players = int(next(i))
-        actions = [int(next(i)) for _ in range(players)]
 
-        nfg = NormalFormGame(actions)
+def to_gam(g, file_path=None):
+    """
+    Write a NormalFormGame to a file in gam format.
 
-        entries = [
-            {
-                tuple(reversed(action_combination)): float(next(i))
-                for action_combination in product(
-                    *[range(a) for a in actions])
-            }
-            for _ in range(players)
-        ]
+    Parameters
+    ----------
+    g : NormalFormGame
 
-        for action_combination in product(*[range(a) for a in actions]):
-            nfg[action_combination] = tuple(
-                entries[p][action_combination] for p in range(players)
-            )
+    file_path : str, optional(default=None)
+        Path to the file to write to. If None, the result is returned as
+        a string.
 
-    return nfg
+    """
+    if file_path is None:
+        return GAMWriter.to_string(g)
+    return GAMWriter.to_file(g, file_path)
diff --git a/quantecon/game_theory/tests/test_game_converters.py b/quantecon/game_theory/tests/test_game_converters.py
new file mode 100644
index 00000000..719fcd5d
--- /dev/null
+++ b/quantecon/game_theory/tests/test_game_converters.py
@@ -0,0 +1,61 @@
+"""
+Tests for game_theory/game_converters.py
+
+"""
+import os
+from tempfile import NamedTemporaryFile
+from numpy.testing import assert_string_equal
+from quantecon.game_theory import NormalFormGame, GAMWriter, to_gam
+
+
+class TestGAMWriter:
+    def setup_method(self):
+        nums_actions = (2, 2, 2)
+        g = NormalFormGame(nums_actions)
+        g[0, 0, 0] = (0, 8, 16)
+        g[1, 0, 0] = (1, 9, 17)
+        g[0, 1, 0] = (2, 10, 18)
+        g[1, 1, 0] = (3, 11, 19)
+        g[0, 0, 1] = (4, 12, 20)
+        g[1, 0, 1] = (5, 13, 21)
+        g[0, 1, 1] = (6, 14, 22)
+        g[1, 1, 1] = (7, 15, 23)
+        self.g = g
+
+        self.s_desired = """\
+3
+2 2 2
+
+0. 1. 2. 3. 4. 5. 6. 7. \
+8. 9. 10. 11. 12. 13. 14. 15. \
+16. 17. 18. 19. 20. 21. 22. 23."""
+
+    def test_to_file(self):
+        with NamedTemporaryFile(delete=False) as tmp_file:
+            temp_path = tmp_file.name
+            GAMWriter.to_file(self.g, temp_path)
+
+        with open(temp_path, 'r') as f:
+            s_actual = f.read()
+        assert_string_equal(s_actual, self.s_desired + '\n')
+
+        os.remove(temp_path)
+
+    def test_to_string(self):
+        s_actual = GAMWriter.to_string(self.g)
+
+        assert_string_equal(s_actual, self.s_desired)
+
+    def test_to_gam(self):
+        s_actual = to_gam(self.g)
+        assert_string_equal(s_actual, self.s_desired)
+
+        with NamedTemporaryFile(delete=False) as tmp_file:
+            temp_path = tmp_file.name
+            to_gam(self.g, temp_path)
+
+        with open(temp_path, 'r') as f:
+            s_actual = f.read()
+        assert_string_equal(s_actual, self.s_desired + '\n')
+
+        os.remove(temp_path)
diff --git a/quantecon/game_theory/tests/test_howson_lcp.py b/quantecon/game_theory/tests/test_howson_lcp.py
index 444dbae4..de30fd5b 100644
--- a/quantecon/game_theory/tests/test_howson_lcp.py
+++ b/quantecon/game_theory/tests/test_howson_lcp.py
@@ -4,7 +4,7 @@
 
 import numpy as np
 from numpy.testing import assert_, assert_allclose
-from quantecon.game_theory.game_converters import qe_nfg_from_gam_file
+from quantecon.game_theory.game_converters import from_gam
 from quantecon.game_theory import (
     Player,
     NormalFormGame,
@@ -22,7 +22,7 @@
 
 def test_polym_lcp_solver_where_solution_is_pure_NE():
     filename = "big_polym.gam"
-    nfg = qe_nfg_from_gam_file(os.path.join(data_dir, filename))
+    nfg = from_gam(os.path.join(data_dir, filename))
     polymg = PolymatrixGame.from_nf(nfg)
     ne = polym_lcp_solver(polymg)
     worked = nfg.is_nash(ne)
@@ -31,7 +31,7 @@ def test_polym_lcp_solver_where_solution_is_pure_NE():
 
 def test_polym_lcp_solver_where_lcp_solver_must_backtrack():
     filename = "triggers_back_case.gam"
-    nfg = qe_nfg_from_gam_file(os.path.join(data_dir, filename))
+    nfg = from_gam(os.path.join(data_dir, filename))
     polymg = PolymatrixGame.from_nf(nfg)
     ne = polym_lcp_solver(polymg)
     worked = nfg.is_nash(ne)
@@ -305,7 +305,7 @@ def test_solves_multiplayer_rps_like():
 
 def test_different_starting():
     filename = "triggers_back_case.gam"
-    nfg = qe_nfg_from_gam_file(os.path.join(data_dir, filename))
+    nfg = from_gam(os.path.join(data_dir, filename))
     polymg = PolymatrixGame.from_nf(nfg)
     starting = [3, 2, 2, 0, 3]
     # We also notice that changing the start
diff --git a/quantecon/game_theory/tests/test_polymatrix_game.py b/quantecon/game_theory/tests/test_polymatrix_game.py
index dc878762..56523b4d 100644
--- a/quantecon/game_theory/tests/test_polymatrix_game.py
+++ b/quantecon/game_theory/tests/test_polymatrix_game.py
@@ -3,7 +3,7 @@
 """
 
 from numpy.testing import assert_, assert_raises
-from quantecon.game_theory.game_converters import qe_nfg_from_gam_file
+from quantecon.game_theory.game_converters import from_gam
 from quantecon.game_theory import NormalFormGame, PolymatrixGame
 from numpy import allclose, zeros
 
@@ -39,13 +39,13 @@ class TestPolymatrixGame():
     @classmethod
     def setup_class(cls):
         filename = "minimum_effort_game.gam"
-        cls.non_pmg = qe_nfg_from_gam_file(
+        cls.non_pmg = from_gam(
             os.path.join(data_dir, filename))
         filename = "big_polym.gam"
-        cls.pmg1 = qe_nfg_from_gam_file(
+        cls.pmg1 = from_gam(
             os.path.join(data_dir, filename))
         filename = "triggers_back_case.gam"
-        cls.pmg2 = qe_nfg_from_gam_file(
+        cls.pmg2 = from_gam(
             os.path.join(data_dir, filename))
         bimatrix = [[(54, 23), (72, 34)],
                     [(92, 32), (34, 36)],