After about two and a half years of studying economics and political science in college, as well as reading a fair amount on my own, here's what I know or believe about economics. I have arguments for everything; if the argument is strong, I know it; if the argument is weak or poorly fleshed out, I just believe it. I'm not going to present the arguments here, just the claims. If you want the argument, you need but ask.
Economics is inherently normative. Anyone who tells you otherwise is trying to sell you something.
Everything in economics is part of one or more dialectical relationships. There are no philosophical "foundations" for anything interesting in economics.
Consequently, the idea of microeconomic foundations for macroeconomics is not coherent. Macroeconomics is its own thing, not just an emergent property of microeconomics. There are an enormous, perhaps infinite, number of ways to conceptualize dialectical relationships between macro and micro.
While dialectical relationships are fundamental, individual micro and macro characteristics and their relationships, can be described statically and linearly. These non-dialectical descriptions, however, are very context sensitive, and describe only where the dialectic happens to be in a specific context at a very specific point in space and time. Generalizing from static descriptions is almost always wrong.
(One useful, albeit deeply imperfect and highly oversimplified, analogy is to a star: the fusing core of the star is the "macro", the outer layers are the "micro". They are distinguishable; both affect each other; neither would exist without the other. Sometimes they're in a dynamic equilibrium; sometimes they're not; sometimes they move from one equilibrium to another; sometimes they explode and/or collapse.)
Macroeconomics (an economy more-or-less as a whole)
By definition, the set of institutions create and implement macroeconomic policy are the government.
It is physically impossible, at the macroeconomic level, to consume more than we produce. We do not have to use public or central economic planning to make sure we do not consume more than we produce.
It is possible and very undesirable to produce more than we consume. If we find that we are (or are about to) produce more than we consume, we should consume more or produce less.
The "exception" to these rules, i.e. international trade deficits or surpluses, are not really exceptions. We can either look at the global economy as the macroeconomy, or we can simply say that without legitimate enforcement mechanisms, and assuming that countries are rational*, a country running a trade surplus has an immediate reason to justify apparently giving away some of its goods and services.
*A dodgy assumption, I'll admit, but dealing with irrationality is not something I know much about.
In an advanced, capital-intensive industrial economy, the "natural" micro rate of investment, i.e. the rate of investment that would occur if there were no specifically macroeconomic institutions, i.e. the government, is negative. In other words, left to themselves, individuals would let existing capital rot until we were no longer an advanced industrial economy.
Except under very restricted conditions that no longer apply in almost every existing economy, the sunk cost of capital cannot be recovered in a perfectly competitive market.
There is a dialectical relationship between the macroeconomic (government) policy and microeconomic (individual) behavior. Neither one "controls" the other, neither can be separated from the other. It is possible to have no macroeconomic institutions, but a capital-intensive industrial economy would then be impossible.
I'm not sure if income and wealth inequality have instrumental effects on macroeconomic aggregates, but distribution is important enough for its own sake that its effects on aggregates is of secondary (but not negligible) importance.
It's useful and unproblematic at the microeconomic level to treat money as just as real as rocks and trees. Money isn't actually real, and it's undesirable at the macroeconomic level to treat it as real, but it must be taken seriously nonetheless. What it means to take money seriously without thinking it's real at a macroeconomic level is so profoundly counter-intuitive that I think one must study economics seriously to avoid grave conceptual errors.
In a capital-intensive industrial economy, it is impossible to even conceptualize, much less measure, the real productivity of an individual or a household*; thus, the concept of an individual household's productivity is not a coherent concept. The income of a household from factors of production (wages, interest, dividends and rents) has nothing whatsoever to do with the incoherent notion of its actual productivity. Whatever does determine (in the loosest sense) a household's income, it is not actual productivity.
*With the possible exception of very few households that do not participate at all in production.
Individual and household productivity was a coherent concept in societies before capital-intensive industrial economics (albeit weakened substantially in intensive agriculture). Most of our intuitions about individual productivity derive from our long history before we were a capital-intensive industrial industrial economy. These intuitions are now not only mistaken, but gravely misleading.
Markets are, conceptually, a good way of distributing some kinds of decision-making. They are a terrible way of distributing other kinds of decision-making, especially certain kinds of Prisoner's Dilemma types of decisions. Markets are a horrible way to make moral decisions.