I'll start by pointing out that I am in no way an economist. Also, while I generally fall into the camp that thinks the minimum wage should be higher than it is, I don't pretend to know what the ideal level is, and I can't be certain that I'm even right that it's too low.
On the smart economists this always bothered me on two fronts. One both groups can't be right. The basic rules of logic mean either raising a minimum wages reduces jobs or it doesn't. It simply can't be both. Saying raising a very low minimum wage a very little may have no measurable effect because of static friction in the labor force but any significant increase will push the needle in the direction of lower jobs.
A simple thought experiment of raising minimum wage to $20 and hour will prove the direction the curve is shaped. The other side of that experiment is eliminating minimum wage. I Dont believe there is any scenario where that would decrease the number of jobs.
I've seen the studies before and I can't imagine any logic where the system behaves any different then I described above. I would truly appreciate you thoughts on this subset as it seems unassailable.
The problem here is, I think people try to apply models that are far too simple to reflect a real economic system. I've always thought economics and meteorology have much in common. They both study absurdly complex systems where the sheer number of variables means that, at best, we can make general statements about what will probably happen over the very short term. For that reason, I see your example as akin to stating that, because I know that the coldest day of the year is generally clear and sunny, while the hottest day of the year usually involves a severe thunderstorm, that as the temperature increases, the severity of the weather must increase. Well, yes, that's generally true, but if we stop there we wouldn't predict blizzards.
For example, any reduction in the minimum wage would necessarily reduce the spending of the workers affected, which in turn reduces demand for the products and services, which then reduce the demand for the workers that produce those products and services. And don't forget, that workers making a little more than minimum wage probably see a reduction if the minimum wage is reduced, expanding the pool of people affected. That's just 1 variable. A economist could easily tick off 100 more just like it. So, to me, it's not hard to see how you could make a logically-valid argument in either direction. And, yes, one is right and one is wrong... at this exact moment. Maybe an increase of the minimum wage from $7 to $7.25 increases employment, but an increase from $8 to $8.25 decreases it. That doesn't seem implausible to me, despite the fact that I have no idea whether its true or not.
Also, not all unemployment is equal. Let's take your $20 an hour. In that case, unemployment would definitely increase, but how much of it would be voluntary? If you take a family where 2 parents work for $11 an hour, and then you boost the minimum wage to $20, maybe one of them quits, even if their job is still available. There are a whole bunch of other reasons why such a move would be terrible for the economy as a whole, but job reduction doesn't tell the whole story.
In studying this I first thought it would take some thinking but then it appeared that what you are saying is that we are going to use the power of the state to make the people having the hardest time getting a job and making a living unemployed. We are going to do this so the people in just a little better position can make more money at the expense of those people we just forced out of the labor market. That doesn't seem very nice.
This assumes that having two people with jobs that don't pay enough to support themselves is better than one person with a job that does pay enough. I would not agree with that assumption. To me, a job that doesn't pay enough to live on is only marginally better than being unemployed, whereas, a job that does pay enough is much, much better than unemployment; so, you may be able to maximize happiness in the universe under the latter scenario. Just yesterday I was listening to an interview with the author of a book that gets into why this might be more true than we realize:
http://www.washingtonpost.com/blogs/wonkblog/wp/2013/09/13/being-poor-changes-your-thinking-about-everything/. In a nutshell, they have found that worrying about money (as low-wage workers often do) taxes our mental "bandwidth" and makes it more difficult to make good decisions and be productive in all aspects of our lives. Kind of like how it's harder to eat healthy when you're feeling shitty (physically or emotionally); because so much "bandwidth" is taken up by dealing with how you feel, your ability to make good food choices is compromised (that's my understanding, not an example from the authors).
In the end though, this:
In your case, apparently you have employees who could take a higher paying job elsewhere but choose to work for you, which to me says you provide a great working environment or some other benefit, tangible or intangible, that they value over higher W2 wages. This illustrates the fact that money isn't everything, and I say good on you!
If you can run your business competitively and genuinely do right by the people who make money for you, I'd say what you're doing is just fine.