No matter how high or how low the market gets, there are always stocks to consider buying and stocks to consider selling.

So today, for the the real-money portfolio I run for the Motley Fool, I've got one buy and one sell.

The sell

By design, my real-money portfolio is banking-centric. That's because banking is the sector I've spent the most time analyzing. However, I'll also buy in other sectors as opportunities present themselves.

I saw such an opportunity in January 2011. Due to budget-related fears of U.S. defense cuts, the defense industry was beaten down; many of the largest players were trading at earnings multiples of around 10 with nice returns on capital.

My thesis was that the largest players in the space would be in a better position than smaller players because, like the big banks, these players as a whole are too big to fail (i.e. the U.S. needs a strong private defense industry for its own protection).

I believed the market was overreacting, but I also believed I could make a misstep if I picked just one company to put all my eggs into. So I bought a basket of five of the largest defense companies trading at low multiples. It was my attempt at making rocket science less tricky.

The worst of the bunch (L-3 Communications (LLL)), has returned 67% before dividends, which roughly kept up with the market's return during that time. The best of the bunch Lockheed Martin (LMT 0.01%) returned considerably better at 159% before dividends. So as a group, the stocks outperformed the market by a good bit.

Before I get into my reasons for selling, let's review the ratios I presented in my buy article back in 2011. First, you'll notice that the P/E ratio of each of the five companies has gone up considerably from 2011 to now (the last two columns):

Company

P/E Ratio (2001 Ave.)

P/E Ratio (2006 Ave.)

P/E Ratio (1/4/2011)

P/E Ratio (Recent)

General Dynamics 

16.6

17.6

10.8

19.7

Lockheed Martin

28.9

17.0

9.8

19.4

Northrop Grumman

12.3

17.1

9.8

15.0

Raytheon

38.3

19.9

10.1

16.6

L-3 Communications

33.5

19.1

9.1

17.2

Source: S&P Capital IQ.

Meanwhile, the returns on capital have been about the same (again, the last two columns):

Company

Return on Capital (2001)

Return on Capital (2006)

Return on Capital (1/4/2011)

Return on Capital (Recent)

General Dynamics

17.2%

13.6%

14.8%

14%

Lockheed Martin

6.2%

17.6%

32.9%

35.1%

Northrop Grumman

6.8%

7.2%

10.5%

12.9%

Raytheon

2.3%

8%

13.3%

12.5%

L-3 Communications

7.8%

8.1%

9.8%

7.2%

Source: S&P Capital IQ.

It's very possible these stocks continue to beat the market from here. As one data point, they're selling as a group at lower multiples than the market currently is and they've certainly seen earnings multiples higher than today's (see the P/E ratios in 2001 and 2006 in the first table).

That said, they're not the screaming buys I thought them to be a few years ago, and, again, defense isn't my area of expertise so I need a larger "knowledge margin of safety" than someone who follows the space closely.

And since these are rather large companies (they range from $10 billion to $60 billion in market cap), I'm less worried that I'm selling a multi-bagger winner.

For all these reasons, I'll be cashing in my position in this basket of defense stocks. But I'll be buying more of a company that is trading near the low earnings multiples the defense stocks were once at.

The buy

I bought into IBM (IBM -0.89%) about a year ago. Back then, I was excited to get a blue chip that was falling on some hard times.

In the interim, things got worse, as reflected by a stock price drop of almost 10%. In its latest quarterly conference call, IBM's CEO Ginni Rometty had to not only announce its 10th straight quarter of year over year sales declines, but also had to lower expectations for 2014 (currently targeting adjusted earnings in the neighborhood of $16 a share vs. the $18 that was once hoped for) and step away from IBM's 2015 target of $20 per share in adjusted earnings.

Long story short, IBM has had a rough go of it as it tries to work the reinvention trick it's pulled for decades now. The IBM of today isn't hardware-focused, but it still relies quite a bit on its profitable software business in addition to its services business (its largest sales driver). As Bloomberg puts it, IBM "struggles to transform fast enough to handle the shift to cloud computing."

Although sales and earnings charts are pointed in the wrong direction, IBM still sits on enough profitability to have a trailing P/E ratio of 13.2 despite taking a $3.4 billion hit as it sells off its semiconductor business. When we look at analyst estimates for next year, we're talking a forward P/E of 9.7 (vs. 17.3 for the market).

I worry less about the services part of the business than the software side, but there are real concerns that this is the time IBM fails to transform itself out of trouble. That said, at these prices (and with a dividend yield of 2.7%), I'm willing to add more to my investment in IBM. It's not too big to fail like I believe the defense contractors to be, but I'm hoping it can follow their path to recovery nonetheless.