Measuring hunger is a challenge indeed, thanks for the interesting blog post.
Whether per capita consumption is derived from food balance sheets or household surveys, both approaches are only proxy measures for hunger, though. They try to measure one key determinant of hunger (food intake), but there are data problems, as you mention, and then intake is only one determinant among others.
Recently I argued that it might be more constructive to look at the actual consequences of hunger when trying to measure it - namely the loss of (physical and mental) functioning that results from undernutrition . Such data is available from Global Burden of Disease studies -- which were pioneered in the Bank's 1993 World Development Report  and which have since then been carried out by the WHO  and more recently in a Gates-fundend project .
This is a bit of a departure from the focus on the "input" side of hunger, but it might be worth giving it a thought.
 Stein A.J. (2013). "Re-thinking the measurement of undernutrition in a broader health context: Should we look at possible causes or actual effects?" IFPRI Discussion Paper 1298. Washington: International Food Policy Research Institute. http://www.ifpri.org/publication/rethinking-measurement-undernutrition-broader-health-context
 World Bank (1993). World Development Report 1993: Investing in Health. Washington: World Bank and Oxford University Press. http://wdronline.worldbank.org/worldbank/a/c.html/world_development_report_1993/abstract/WB.0-1952-0890-0.abstract1
 WHO (2014). "Global Burden of Disease (GBD)." Website. Geneva: World Health Organization. http://www.who.int/healthinfo/global_burden_disease/gbd/en/index.html
 IHME (2014). "Global Burden of Diseases, Injuries, and Risk Factors Study 2010 (GBD 2010)." Website. Seattle: Institute for Health Metrics and Evaluation. http://www.healthmetricsandevaluation.org/gbd/research/project/global-burden-diseases-injuries-and-risk-factors-study-2010