Not true. Austrian's acknowledge that models can be useful (and have used them) for pedagogical purposes, they just aren't very useful for predictions.
I think Rothbard put it best:
Here is the problem for the Neo-Austrian/Neo-Rothbardian anarcho-capitalist; while rejecting models for lack of realism, the perfect competition model IS the anarcho-capitalist theory of society. This, of course, is highly problematic.
Disagree as well. The Anarcho-Capitalist position is that no government is superior to any government, which relies in part on economic analysis but is not wholly dependent on a specific model.
The anarcho-capitalist position is not derived from the naïve perfect competition model (which is, as a model construct, useful for pedagogy, but not so when attempting to obtain a more nuanced understanding of the market). In fact, to get to the anarcho-capitalist position from a mainstream economic model one must simply consider the following:
If you recall, Pareto efficiency means that no change to the situation can be made that isn’t at the expense of any particular individual. Furthermore, economies tend towards equilibria over time due to trade, which is by definition mutually beneficial.
Therefore, a competitive market will result in improvements in the quality of life for all participants at the expense of nobody (because of trade), and will tend towards an equilibrium that is efficient and where no central planner could possibly improve the situation (because of the First Welfare Theorem), which is tautologically true under certain conditions.
The primary economic criticism of this position is the assumptions built-in to what makes a market competitive and why trade is mutually beneficial. The former is usually expressed in terms of various points of “market failure”. But criticizing the market for not being perfect is one thing; the question is, is it possible to create a sustainable system to address market failure?
The first criticism, that market failure can arise via imperfect information and therefore government is needed, is absurd. Perfect information can be safely ignored without affecting an argument in favor of minimal/no government. This is because both government and private actors face the same information constraints. Information, in a sense, can be (and is, in Austrian economics) no different from any other good and service. Consumers with superior information make more informed choices which leads to better gains from trade, just like producers with superior capital have lower costs which leads to better gains from trade. There is an incentive to accumulate valuable, practical information in the same manner as there is an incentive to accumulate capital.
Economic accumulation of useful information leads to superior quality of life for the person in question. Similarly, when it is not worth the cost to gather relevant information, the tendency to do so unnecessarily will be curbed. As such information (or lack of it) is not a significant constraint on the market except to say that there are always improvements to be made and room for entrepreneurship. This dynamism is what makes free markets a system and not a state; it has built-in tendencies towards resolving this problem over time since it rewards entrepreneurs who act on these information disparities.
Comparatively, there are less, and weaker, incentives for politicians, voters, and bureaucrats to be as informed, as Public Choice theory explains. Big Government types invariably carry with them the belief that they know what’s good for you and can fix all the problems; in some (coincidental) cases they might, but there is no reason to believe it is a systemic feature of government (in fact, there is ample reason to believe otherwise when examining history). People reason that because it’s possible for individuals to criticize a particular state of the market at a particular time (even a seemingly persistent state), this justifies inserting a system of government over top of it. This is unfounded. Those who propose such arguments rarely have a coherent and well-designed “system” in mind, only various states they wish to resolve through which they attempt to justify their impractical system. The failures of the government system are legendary but no pro-government type has taken sufficient steps to resolve this.
Similarly, the “complete markets” assumption can safely be dropped when advocating for a pro-market position. The complete markets assumption basically boils down to the fact that to the extent that something isn’t traded on the market (particularly as it pertains to future speculations), calculations involving it will tend to be suboptimal and externalities will arise where one group can benefit at the expense of another. The Austrian response to this is to eliminate said externalities by expanding the market to encompass such things, thereby resolving the problem.
The Statist response is to expand the role of government. This is even less defendable than the prior position. Whereas the preceding one is framed as an empirical or pragmatic position (“well if you just elect the right people to enact the right policies…”), this one is a solely deductive critique that fails to make any coherent sense. In effect, it states: “The market fails to work efficiently when its actors cannot make wholly internalized profit-and-loss calculations over certain goods and services. We fix this problem by making it harder, impossible, or illegal for actors to make wholly internalized calculations, since we hand the job over to an external party thereby magnifying the problems of externalization.” The government is just as susceptible to critiques of externalization, given that a given government policy is generally the literal embodiment of externalized costs and concentrated benefits.