You really think the perpetual growth rate is low enough (or even negative?) to make XOM's total return over the next ten years worse than that of a 10-yr. Treasury? The dominant secular trend impacting this assumption includes crude becoming more expensive to get out of the ground, whereas the dominant cyclical consideration is global GDP. Even with XTO in the business mix, it seems like a stretch to believe the cyclical driver will dominate the secular driver over a ten year period.
Even if you're right, C/(r-g) gives you just one valuation data point, and the market uses several (e.g., comps for price / 2011 earnings are 6.9x, 10.0x, 7.3x, 4.6x and 7.4x for XOM, CVX, TOT, PBR and COP, respectively, per Bloomberg's median). Based on comps, I would think the principal appreciation potential of XOM would be greater than that of the stretched 10-yr. Treasury.
Your explanation seems like it could be only a part of the story..