Investors' lust for dividend stocks is driving up prices and making a simple 3% return harder to find. So MoneyShow's Jim Jubak, who also writes for Jubak's Picks, is replacing three picks in his Dividend Income portfolio.

In the 0% interest-rate world of Ben Bernanke, the 3% dividend yield is king.

When a two-year Treasury note yields 0.22% and a two-year CD pays 0.85%, it's not surprising that savers and investors are eager to snap up anything with a higher yield.

That's got an upside—stocks that pay 3% or more have shown big gains in price as dividend-hungry investors have bought the shares. Intel (INTC), for example, which paid a 3% dividend at the end of 2010, returned 19.03% in 2011 (combining price appreciation and dividends.)

And it's got a downside—as investors pile into a stock yielding 3% or more, the dividend yield goes down as the price goes up—even if the company increases its dividend payout. Intel paid out 78 cents a share in dividends in 2011, versus 63 cents in 2010, but thanks to the climb in the stock's price, the yield has fallen below 3%—to 2.94%, as of February 1.

Companies recognize this hunger and, as I wrote in my November 24 column, "6 Companies with Soaring Dividends," they're aggressively raising dividends, because they realize that in the current low-yield world it's an extremely effective way to support stock prices.

Right now, investors prefer a higher dividend to a share buyback. This has led companies to shockingly hefty dividend increases.

One recent example is Mattel (MAT), which lifted its dividend 35% on January 31. A full year's payout at the new rate works out to a yield of 4% on the January 31 closing price, even after the share price jumped 5% on the news.

And I don't think the trend pushing up the price of stocks yielding 3% or more is about to come to a quick end.

At its January 25 meeting, the Federal Reserve's Open Market Committee said it would keep short-term interest rates at their current exceptionally low levels—I guess 0% counts as exceptionally low—until the end of 2014. That's a big extension of the Fed's previous guidance for interest rates at this level until mid-2013.

That poses an interesting problem for savers and investors. We all want higher yields, and we certainly don't mind cashing in on any price appreciation. But the appreciation in share prices constantly pushes the yield down on these stocks.

That's not a problem for investors who already own shares. They locked in their yields when they bought. But it is a problem for investors with new money, as yesterday's high-dividend stock turns into tomorrow's stock with a mediocre yield.

And settling for a declining yield because we locked in a good yield when we bought our shares a year or two or three ago strikes me as passing up one of the best things about the current dividend craze.

Many companies now see a higher-than-3% yield as the best way to support their share price in a sometimes-difficult market, so many companies are working hard at raising their dividend payout fast enough to keep pace with their rising share prices.

The goal is to add enough to the dividend payout every year (or even every quarter) to keep that yield above 3%, even if the share price is climbing. So, for example, Intel—which saw its yield slip below 3% as its share price climbed in the first half of 2011—upped its quarterly dividend to 21 cents from 18.12 cents with the August quarter.

Of course, not every company has the cash to do that—or a management that's committed to increasing the dividend at that pace. But as a saver or investor in this financial environment, you'd sure like to have more (rather than less) of those stocks in your dividend-income portfolio.

So that's why I'm going to do a fine-tuning of my Dividend Income portfolio today. This won't be my annual full-scale performance report and portfolio overhaul. I'll keep that for May, as usual.

But I have gone through the portfolio looking for companies that I think might fall behind in the dividend race, and I've replaced three of the picks that concern me on that basis with new dividend plays.

Up Next: So which stocks go?...

|pagebreak|

So which stocks go? Just three for right now.

Abbott Laboratories (ABT)
When the company reported earnings on January 25, it announced that it would resume its share-repurchase program. But it left the dividend at 48 cents a share, where it's been since April 2011. (The most recent quarterly dividend was paid to shareholders of record as of January 13.)

My suspicion is that the company is wrapped up in plans to split into two companies by the end of this year. I don't think investors can expect a significant dividend increase before the split.

The current dividend yield of 3.54% will stay above 3% only as long as the stock doesn't appreciate significantly. That's not the combination I'm looking for in this dividend-happy stock market.

Merck (MRK)
Merck increased its quarterly dividend to 42 cents a share from 38 cents in the last quarter of 2011. That pushed the stock's current yield on a 2012 dividend of $1.68 to 4.3%, at the February 1 price of $38.63.

But I think that's Merck's last dividend increase for 2012. In August 2012, the company loses patent protection on asthma and allergy drug Singulair, and Singulair represents 10% of Merck's sales.

The company has a solid pipeline of new drugs that should be able to make up for the loss of Singulair in 2013, but there's a good chance the company will report a single-digit decline in sales for 2012. That's the kind of uncertainty that keeps a board of directors from raising dividends.

Potlatch (PCH)
The current depression in the housing industry makes these tough times for timber companies. Potlatch actually cut its dividend payout from $2.04 in 2010 to $1.84 in 2011.

Dividends are projected to drop even lower in 2012, since the company reduced its quarterly payout to 31 cents a share in December from the previous 51 cents. That brings the projected yield for 2012 down to 4.08%, from the current trailing dividend yield of 6.06%. That's the wrong direction.

Although I think we're starting to see signs of a bottom in homebuilding, I don't think the recovery will be quick enough to add to Potlatch's dividend payout in 2012.

My gains on Potlatch since I added these shares to my Dividend Income portfolio on December 9, 2008 are 37.65% (plus an original dividend yield of 7.2%). My gains on Abbott Laboratories of 3.3% and on Merck of 6.5% are much smaller, since I added these stocks to the Dividend Income portfolio on May 6, 2011. The original yields on those two stocks were 3.4% and 4.15%, respectively.

Then, for my May annual revision, I'm keeping my eye on two other stocks currently in the portfolio, French oil company Total (TOT) and US coal-mining master limited partnership Penn Virginia Resource Partners (PVR).

Both still have projected yields for 2012 high enough—4.73% and 7.83%, respectively—to make them tough to replace. But I worry about the pressures that falling energy demand in a slowing global economy are putting on cash flow.

3 Dividend Plays to Add Now
What to replace them with? Why, companies that are showing good prospects for future earnings growth that will lead to higher dividends, of course. (And are already paying out a 3%-plus yield.)

Such as?

Kinder Morgan Energy Partners (KMP)
The partnership paid $4.32 a unit in 2010 and $4.58 in 2011, and, thanks to new pipelines serving the US energy boom and the likely drop-down of assets from general partner Kinder Morgan's (KMI) acquisition of El Paso (EP), I think the partnership will see growing cash flows that it can pass through to unit holders.

(Master limited partnerships aren't suitable for tax-sheltered retirement accounts in all circumstances. Please check with your accountant to see if a purchase might put you over the income limits for this kind of account.)

The current yield on these units was 5.38% on February 1.

Westpac Banking (WBK)
The Australian bank grew dividends to $8.06 in 2011, from $6.41 in 2010. The current trailing 12-month yield is 7.12%. Like many overseas companies, Westpac doesn't pay quarterly dividends, instead making its payouts twice a year, in May and November.

Australia's banking market is divided among just four major banks, which gives Westpac important advantages of scale (and a lack of outside competition). The biggest worry is the Australian housing market.

I think tight underwriting and low loan-to-valuation mortgages make this a very different market from those in the United States and the United Kingdom. But if you're worried about an Australian housing bubble, this isn't the dividend stock for you. (Westpac Banking is already a member of my Jubak's Picks portfolio.)

General Electric (GE)
GE increased its dividend twice in 2011 after upping it twice in 2010. The December increase brought the projected 2012 yield to 3.62%. That's not especially juicy, but I'm betting that General Electric will continue to add to its dividend payout in 2012.

In the company's fourth-quarter conference call, CFO Keith Sherin told investors not to expect the company to use all of its current $12 billion in excess cash on dividend payouts. The company expects to keep $8 billion in reserve and use the other $4 billion for acquisitions and dividends.

Add in cash flow from operations, and there's clearly enough in the vault to fund another couple of dividend increases in 2012—especially if the US economy keeps chugging along at its surprisingly strong (if modest) pace.

I'll be making these adds and drops to my Dividend Income portfolio over the next few days.

Full disclosure: I don’t own shares of any of the companies mentioned in this post in my personal portfolio. The mutual fund I manage, Jubak Global Equity Fund, may or may not now own positions in any stock mentioned in this post. The fund did own shares of Abbott Laboratories and Westpac Banking as of the end of September. For a full list of the stocks in the fund as of the end of September see the fund’s portfolio here.