This calculation has been done many times, and is kinda silly for a number of reasons.
First, when people use the term, "size of the economy" what they're referring to is GDP, which is a measure of how much money was spent that year. GDP itself is a frightfully terrible way to measure economies, but we use it because governments like to show that when they spend money, the economy grows. Any "valuation" of Bitcoin derived from any GDP figures is already going to be pretty useless.
Second, the GDP figure measures spending, but most monetary units spent during the year are spent numerous times. If GDP is $1 trillion, and each dollar is used ten times, then "valuing" the dollars would more appropriately be $1 billion/number of dollars in existence. Your equation completely ignores this, and it's extremely relevant. I believe the proper term here is "monetary velocity." There are also issues of fractional reserve, credit expansion, and derivative instruments which change the supply of money far beyond its actual supply, and much of this would still occur under a Bitcoin-based monetary system.
Third, none of the above to criticisms even matter, because the main problem with this equation is that the value of something cannot be determined solely by seeing how much it is used. The value of any item, whether it's cows, cars, dollars, or bitcoins, derives from numerous factors which are typically distilled down to the phrase "supply and demand." How widely something is used influences this, but it's only a part of the valuation and not even the majority.
In other words, it's fun to make that calculation, and attach a figure of dollars to what a Bitcoin "would be worth if everyone used it." But it's a completely fallacious figure, and if indeed everyone used Bitcoins only, I promise the price would be far higher, or far lower, than your estimate, which makes it not a very good estimate at all
