As global economic uncertainty continues to erode investor confidence, bond mutual funds are getting the lion's share of new investment dollars. So what's the best play in bonds?

I'll go into detail in a moment, but here's the upshot: Even though you might be seeking safety by investing in bonds rather than stocks, the best way to play the bond market is by taking a little risk.

First, there's one thing about the bond market that you need to know. It's just my opinion, but I believe it so strongly that I treat it in my own mind as an established fact. The most critical element in bonds is interest rates -- when rates go up, bond prices go down, and vice versa. The most critical element in interest rates is Federal Reserve policy. In my judgment, the Fed is definitely not going to raise interest rates from where they are now -- essentially zero -- for the rest of the year. Very probably they will be essentially zero all next year, too.

This means that you should be willing to take the risk in the bond market. That means you shouldn't be afraid to buy long-term bonds. They are the ones that have the highest yields, but will get hurt more if rates rise. For example, now a 30-year Treasury bond has a yield of more than 4%, and a 10-year has a yield of about 3.2% At the same time, a two-year only has a yield of less than 0.7%

Purely as a short-term trading matter, I think yields on long-term bonds right now are a touch low, and I think you could get higher yields and better prices if you wait a couple weeks or months. But that's just tactics. Strategically, unless the Fed raises its policy rates, there isn't going to be a gigantic move upward in Treasury yields.

It also means that you should be willing to take some default risk in the bond market. That means instead of buying Treasury bonds you should buy corporate bonds, even high-yield or so-called "junk" bonds. The reward for doing so is clear enough. You can get more than 9% on five-year high-yield corporate, but less than 2.0% on a five-year Treasury.

Yes, the "junk" bonds can default if the issuing company gets into trouble, while at least in theory the U.S. Treasury can never default. But if I'm right, and the Fed isn't going to raise interest rates for over a year, that means the economy is going to be awash in liquidity, and very few companies are going to default. Even in the worst of the credit crisis of 2008 and 2009 very few high-yield bond issuers defaulted. If you'd invested in a diversified high-yield bond mutual fund with hundreds of bonds in its portfolio, you'd never even notice if one or two went under. The reason so few defaulted was because the Fed stepped in and saved the banking system with zero interest rates -- and that's just what the Fed is going to continue doing.

How can I be so sure?

It's actually quite simple. The Fed is an arm of government that is bound by laws about what it can and can't do, and that define its mission and purpose. The Fed is tasked with the dual mandate of ensuring "price stability" -- that is, no inflation or deflation -- and "maximum employment" -- that is, jobs.

Right now inflation is very, very low. In fact, it's so low that it's almost deflation. The Fed fights inflation with high rates, and deflation with low rates. So right now, it's an open-and-shut case for low rates as far as "price stability" is concerned.

Right now the unemployment rate is egregiously high, almost 10%. The Fed fights unemployment with low rates. Again, an open-and-shut case for continued low rates as far as "maximum employment" is concerned.

The Fed is actually amazingly predictable. An economist at the Federal Reserve Bank of San Francisco has come up with a simple rule based on inflation and unemployment. It pretty much perfectly explains the Fed's interest rate policies going back almost a quarter century. You just tell the rule what the inflation rate and the unemployment rate are, and it will tell you what interest rate the Fed will set.

If you put in today's interest rate and today's unemployment rate, guess what interest rate the rule says the Fed will set? Today's rate of zero? Uh, no the rule says -6%!

For the first time, the Fed simply can't follow the rule because it can't set rates at, or below, zero. So with rates jammed down as far as they can go, the Fed has done other things to make up for it -- such as buying more than a trillion dollars of mortgage-backed securities. The same Fed economist estimates that makes today's zero rate actually something like -2%. But that's still a long way from -6%.

That means that, if anything, the Fed is going to find more ways to effectively lower interest rates even more, by hook or by crook. It certainly won't do anything to raise them, until and unless the inflation rate rises and the unemployment rate falls.

Well, you've seen the latest numbers. Last month's jobs report has the unemployment rate still at 9.7%, and the only reason it's that low is because the government hired a 400,000-plus temporary census workers. And

Wednesday's inflation report showed that the CPI was actually negative last month.

So do you see what I mean when I say that this is just my opinion, but that I nevertheless treat it as fact? Based on this, just give me one good reason why the Fed would even remotely consider raising rates? I don't even know why they'd bother to have a meeting next week. Maybe I do -- it could be their chance to talk about lowering rates somehow.

If you agree with this analysis of the Fed, then the bond strategy is clear. Buy long-term bonds, and buy risky bonds.

What does it say for stocks? That's not so obvious, because stocks are influenced by a lot more than interest rates. But knowing that the Fed is going stay at a zero rate -- effectively, even lower -- for more than a year means that the economy is going to continue to get ongoing stimulus. If nothing else, that means the stock market correction we're in now is likely to just be a correction, nothing more.