Self-interest certainly plays a role.  For one thing, banks and financial companies employ the most economists and pay the best salaries.  That fact has an enormous influence on the axioms that economists choose, e.g. what should be the goal of economic planning and regulation, and what is taught at economics schools.  Even those economists who reject those axioms choose their own axioms based on their political preferences or business interests.  How could it be otherwise?

Lack of data is also another problem.  The world's economy is such a complicated system that no one can know its state and workings well enough to make any useful predictions.  Technological innovation is a major source of uncertainty. So are natural events and disasters (wars, droughts, eruptions, oilfield discoveries, nuclear meltdowns, etc.)

Yet another problem is that few if any economists have the mathematical knowledge and tools that are needed to usefully model a system of that size.

For example, some very crude models of the economy start assuming a linear feedback between, say, supply or price of each major product (oil, coal, electricity, grain, etc.) on the price of other products.  Mathematically, such a model includes a matrix of coefficients with tens of thousands of rows and columns, that express those mutual influences.    So, even if economists could get the data to fill that matrix, they would be unable to use that model for predictions or insight.  They would probably do what they do now: pick a few products, guess a few entries in that matrix, and pretend that they can make useful predictions from them alone, without considering the rest of the matrix. But one of the most important things one learns in linear algebra is that one cannot use intuition or shortcuts to predict the behavior of such linear feedback systems: small changes in just one of those millions of numbers could make the system behave in completely different ways.