Tobin Q Data

March 23, 1999

This month I'd like to present another way of looking at aggregate market levels besides the traditional PE ratios, price/book value, price/sales ratio, dividend yields and the other more commonly used approaches.

Nobel Prize winning economist, Dr. James Tobin, devised a concept called "Q" or "Tobin's Q". Q is simply the ratio of the market value of an asset divided by its replacement cost. When applied to individual businesses, Q does not seem to be particularly useful. Many companies posses competitive advantages or terrific reputations that enable them to enjoy above average returns on capital on a sustainable basis. This goodwill makes them worth significantly more than the value of their assets. Similarly, there are some companies that earn very low average returns. They are therefore worth less to their owners as a going concern than the value of their assets.

However, on an aggregate basis, Q has tended to be mean reverting. There has been a tendency for the aggregate market value of stocks and the aggregate value of their assets to converge at 1.0 (or 100% in the table that follows). More interesting is that the Q ratio is often far from its mean level depending on the general economic environment. Therefore it has been especially useful in making longer term value judgments. I should add that it is important to understand that Q is NOT a timing tool. It is a valuation tool. It will only help an investor find generally attractive or unattractive markets from a valuation perspective.

The Federal Reserve provides aggregate replacement and historical costs for United States business in its Flow of Funds Reports. The data is for both public and private, non-farm, corporate, non-financial businesses. Unfortunately, we have market values for public companies only. The data has also gone through several updates. The Fed is constantly trying to improve the quality of its reports. But despite these problems and the slight mismatch in the two series, Q has still been a remarkably useful gauge of aggregate stock market levels from the 1920s until the present.

Q reached levels that were similar today's (but not quite as high) in the late 20s and middle 30s. Both periods were followed by serious market declines to levels based on Q and subsequent market action were very attractive entry points.

Q correctly signaled under-valuation in the late 40s and early 50s. A period that proved to be an excellent entry point.

Q signaled significant over-valuation in the late 60's and early 70s. This period was followed by the significant market break of 1973-74 which based on Q and subsequent market movements was a very attractive market.

Q again correctly signaled a terrific entry point in the late 70's and early 80s.

Q is now at its all time high and signaling massive overvaluation. It remains to be seen.....

The following table contains the Federal Reserve data from 1945 through 1997. The mean 'Q' for this period was about 60%. The 60% (as opposed to 100%) is primarily the result of the lack of market value information for private businesses. The last column is adjusted to that mean and tells us the degree of under/overvaluation in clearer terms.

Andrew Smithers and Stephen Wright of Smithers and Co. Ltd. did the extensive research on the "Q" data from 1925 through 1945 that is not included in the table.

 

Year Replace Historic Market Q Ratio Q Ratio/
  Cost Cost Value   Mean
           
45 195.9 126.4 103.7 52.94% 89.33%
46 222.3 133.4 97.3 43.77% 73.86%
47 261.7 140.3 95.1 36.34% 61.32%
48 282.7 152.3 94.1 33.29% 56.17%
49 292.8 162.0 103.8 35.45% 59.82%
50 321.2 175.7 126.7 39.45% 66.57%
51 352.2 193.0 147.6 41.91% 70.72%
52 369.9 207.0 146.6 39.63% 66.88%
53 383.7 220.6 144.1 37.56% 63.38%
54 395.3 232.9 194.8 49.28% 83.16%
55 426.6 249.5 244.2 57.24% 96.60%
56 468.5 268.3 268.4 57.29% 96.68%
57 493.3 284.6 245.7 49.81% 84.05%
58 504.6 300.0 324.7 64.35% 108.59%
59 525.6 320.8 362.2 68.91% 116.29%
60 533.5 334.3 365.2 68.45% 115.52%
61 548.3 354.7 437.7 79.83% 134.71%
62 566.1 375.2 424.3 74.95% 126.48%
63 584.2 395.6 465.8 79.73% 134.55%
64 610.3 416.5 546.0 89.46% 150.97%
65 650.4 451.7 623.8 95.91% 161.85%
66 697.4 483.4 547.9 78.56% 132.58%
67 752.2 516.9 712.2 94.68% 159.78%
68 814.3 549.9 843.2 103.55% 174.74%
69 891.9 593.5 694.6 77.88% 131.42%
70 970.5 622.4 677.0 69.76% 117.72%
71 1066.3 664.4 792.9 74.36% 125.49%
72 1186.0 742.2 994.3 83.84% 141.48%
73 1344.9 809.4 759.2 56.45% 95.26%
74 1692.3 975.1 510.7 30.18% 50.93%
75 1731.8 872.0 699.1 40.37% 68.12%
76 1929.4 963.3 848.3 43.97% 74.20%
77 2141.9 1050.7 740.4 34.57% 58.33%
78 2415.8 1161.5 777.3 32.18% 54.30%
79 2762.3 1288.4 922.5 33.40% 56.36%
80 3174.9 1421.3 1214.6 38.26% 64.56%
81 3524.2 1573.5 1110.7 31.52% 53.19%
82 3691.5 1675.2 1264.4 34.25% 57.80%
83 3853.1 1863.8 1486.1 38.57% 65.09%
84 4084.2 2063.1 1370.7 33.56% 56.64%
85 4194.1 2182.8 1674.7 39.93% 67.38%
86 4339.6 2307.1 1978.1 45.58% 76.92%
87 4568.9 2459.0 2014.6 44.09% 74.41%
88 4882.3 2619.6 2216.7 45.40% 76.62%
89 5088.1 2686.9 2673.1 52.54% 88.66%
90 5099.0 2775.8 2529.7 49.61% 83.72%
91 4977.7 2921.9 3497.0 70.25% 118.56%
92 4683.2 2969.8 3827.9 81.74% 137.93%
93 4758.0 3238.6 4126.0 86.72% 146.34%
94 5147.9 3513.7 4141.8 80.46% 135.77%
95 5572.2 3872.6 5481.1 98.37% 166.00%
96 6013.1 4306.4 6367.4 105.89% 178.70%
97 6797.5 4821.1 7793.8 114.66% 193.49%

The adjusted Q ratio at the end of 1997 was 193%. Considering the significant market rise since then, stocks would appear to be selling for more than twice their long-term average and fair value based on Q. When I was confronted with yet another indication of significant overvaluation I decided to write to Dr. Tobin and ask him what he thought about the current high level of Q. I specifically asked him:

Did he think there would be a crash?

Did he think that the current high return on capital that theoretically justifies higher stock prices was sustainable?

Did he have any other insights to share?

I am pleased to report that not only did he reply, he agreed to let me share his thoughts on the subject with you.

Dear Mr Crimi,

I am agnostic. Many people do conclude from the record high value of Q that the market value of equity is bound to fall. They may be right, though they haven't been yet. There are other logical possibilities. Maybe there has been a fundamental reduction in the cost of equity capital, the risk premium on stocks relative to bonds and federal funds. Then the high value of Q is a signal and incentive for real productive business capital investment; gradually the increase in stock of corporate capital at replacement cost will reduce the marginal productivity of capital and reduce Q , but no bubble and no crash. Another problem is that increasingly the capital value of companies is not tangible assets but human capital, more precisely the ability of the management to find and hold the brightest innovators. It's like betting on the coach of a sports team. That kind of asset is not in the denominator of Q as we calculate it.

Dr. James Tobin

If I understand him correctly, Dr. Tobin holds the view that stocks are expensive, but not quite as expensive as they look based on Q. He is also not sure how the overvaluation is going to resolve itself. It may resolve itself in a crash. It may resolve itself in an extended period of tangible investment until return on capital falls, asset values rise, and the stock market treads water.

In any event, this suggests that stocks do not offer very much in the way of decent returns over the coming years. Incidentally, I'd like to formally cast my vote for a crash.

China has only 2% of its Total Foreign Reserves in gold.