This is a point easiest to explain by counter-example (if "small is sustainable", then is large unsustainable?). Consider the national highway system. It took (and is taking) the federal government hundreds of billions of dollars to build and maintain that monstrosity. Only a large, centralized body with access to vast resources (and apparently untapped reserves of hubris) could have conceived of something like that, let alone have built it. And the building of it is one of the many drivers that has set our nation, literally, on the road to ruin.
*I am uninterested, at this time, in defining the word "sustainable."
So: small is sustainable. With limited resources, one is forced to be creative, resourceful... small. If a mistake is made, terrifying amounts of wealth are not extinguished thereby; it remains a mistake, rather than a catastrophe: one is free to learn and try again. If one succeeds, then others, also of limited means, may learn of that good practice, adapt it to their particular circumstances, and try it themselves. Their success -- or failure -- contributes to the breadth and depth of our collective well of wisdom.
Many small groups experimenting is much better than one centralized entity pooling resources from a large area and rolling the dice. I would not suggest it is never advisable to pool resources, but I am asserting that this ought to be the exception rather than the rule. And just because I'm a hopeless romantic who loves lost causes, here's one exception: universal, single-payer healthcare. This would support small solutions by granting working people and entrepreneurs the freedom to try and fail and move on if and when they choose; there would no longer be a need to stay in a job out of fear of losing all-important health coverage. This enables small enterprise in another way: it lifts the burden of one of the heaviest costs of business, which is caring for employees' health. This is but one example of risk management which, in general, is a sector that benefits from the pooling together of large groups.
Smallness is resilient. The failure of one does not precipitate the failure of another; to the contrary, it may stimulate their improvement by the example of what not to do. Further, the failed, having committed relatively few resources, and presumably surrounded by non-failures, may recover with less delay. Smallness also, by nature, is less interdependent (and more, but not completely, independent). Smallness can't draw on limitless territories for natural and human capital: it must make do with what is close at hand. This contributes to the diversity of experimental solution-finding, but also limits the small's vulnerability to drawn-out resource chains. Consider our situation today, when, by a hilariously perverse (perversely hilarious?) turn of events, the Pentagon finds itself reliant on Chinese rare-earths for key components to its "advanced" weapons systems. Not so "advanced" when they rely on the goodwill of a creditor and powerful economic competitor, eh? Smallness doesn't have that problem... and not least because it probably wouldn't be in the habit of dropping "smart" bombs on some faraway desert to secure the oil it doesn't need.
So, smallness is resilient. But what about the case of systemic failure? This would most likely be the result of neglecting the advice above and tending too far in the direction of centralization of resources and power. As this seems to be where we're trending, however, the question may be worth essaying an answer: in the case of systemic failure, save what's worth saving, discard what's not, and start again. It wouldn't be the first time.