The young black doctors who earned the same salary as we whites could not achieve the same standard of living for a very simple reason: they had an immense number of social obligations to fulfill. They were expected to provide for an ever expanding circle of family members (some of whom may have invested in their education) and people from their village, tribe, and province. An income that allowed a white to live like a lord because of a lack of such obligations scarcely raised a black above the level of his family. Mere equality of salary, therefore, was quite insufficient to procure for them the standard of living that they saw the whites had and that it was only human nature for them to desire—and believe themselves entitled to, on account of the superior talent that had allowed them to raise themselves above their fellows. In fact, a salary a thousand times as great would hardly have been sufficient to procure it: for their social obligations increased pari passu with their incomes...
The thick network of social obligations explains why, while it would have been out of the question to bribe most Rhodesian bureaucrats, yet in only a few years it would have been out of the question not to try to bribe most Zimbabwean ones, whose relatives would have condemned them for failing to obtain on their behalf all the advantages their official opportunities might provide. Thus do the very same tasks in the very same offices carried out by people of different cultural and social backgrounds result in very different outcomes.
Now, I'm no Africa expert, but clearly Africa is a basket case (The term 'basket case' came from WWI, indicating a soldier missing both his arms and legs who needed to be literally carried around in a basket). To the extent this social dynamic, where individuals cannot become wealthy because they have so many contemporaneous obligations they can never accumulate savings, has serious implications for a growing economy. After all, most small business is funded via retained earnings, and if that is all going out to one's extended family, that won't happen.
Now consider if this is true and important. Macroeconomic variables will not capture it. This factor may not be operative, but it's plausible, and perhaps there's another such theory at work that is not captured by national income accounting or unemployment statistics. Macro thus appears like measuring someone's health using a thermometer, noting dead people are cold, sick people are hot, and healthy people are 98F (36C). But that metric is not very helpful in predicting, or even explaining, people's health.