I'd agree that your numbers are the best available. My problem is with coming up with output figures for service work. Manufacturing is easy it figure. Take shoes, the BEA has data for all sorts of different kinds of shoes. For restaurants they lump Starbucks in with Four Seasons. Quality is just one problem for the bean counters; darn near everything is a headache in figuring output for services.
A shoe factory makes shoes and you can count them. Someone could say that output from a trucking company equals freight tons X miles. They'd be screwing up because we might hire some computer geek to reroute our trucks to make all deliveries with less driving. The bean counter will say my output (freight ton miles) is falling while my banker will say that I'm getting rich. For me, the most consistent way of measuring output in the service sector is by looking at profits, which is why I say the transition from manufacturing to services is good for America.
Here's the problem with a service business: when you stop performing services, the income stops. If you have mining, or farming, or manufacturing, you can hire the productive work and the income continues on.
For example, how much better off is the guy or gal who makes $ 175,000 in dividends and interest ( or other passive income ) than the lawyer or physician who works every day to earn an equal income? When the lawyer or physician takes a month off the income stops.