One of the biggest problems with newspapers is that most of the people who write for them have the attention span of a flea, and act as if readers are that way, too. We breathlessly tell you today's news, but we rarely go back and tell you what happened to yesterday's news. And we almost never admit when we've made a mistake.

In honor of the new year -- and because it's easier to look at old articles than to find new things to write about -- I would like to go over 1990 with you. Especially the times that I had the facts right, but drew the wrong conclusion from them.

I had a good batting average in 1990, because my stick-in-the-mud kind of thinking came back into fashion. I think it's foolish to borrow more money than you can pay back. I also think that if a person, a corporation or a country is always a borrower and never runs a cash surplus, the lenders will catch on one day and cut off the money. In the 1980s, especially the late 1980s, these beliefs made me a dinosaur. In 1990, they made me a prophet.

My three best predictions of the year -- the collapse of the junk bond market, the fall of Donald Trump and the bankruptcy of the Frank Lorenzo airline empire -- were all simple logic, if you think the way I do.

Early in 1990, anyone with two eyes, half a brain and an open mind could see that the junk bond market, in its late 1980s form, was finished. Dead, kaput. In the 1980s junk was a Tinker Bell market -- everyone clapped and believed in junk bonds, so there was a market, even though many junk issuers were financially unsound. When Michael Milken was driven out and everyone finally stopped believing, you had a market in which you could buy all the junk bonds you wanted, you just couldn't sell them at a reasonable price. Junk bonds, to quote one of my February columns, became a financial Roach Motel. You can check in, but you can't check out.

That made calling the fall of Donald Trump easy. Especially after Forbes magazine, in April, published Trump's balance sheet. Do a little forensic accounting, and you saw that Trump had almost no cash cushion and his properties weren't producing enough cash to pay his interest bills, and that the only way he would emerge with anything other than his mouth is by making deals with his lenders, those poor souls. I've missed a prediction or two on The Donald, but in all, my record on Trump's deals is lots better than Trump's is.

I made the Lorenzo call with help from my Newsday colleague Glenn Kessler, who covered the Eastern Airlines bankruptcy. I realized that the deficit in Eastern's pension fund could sink Lorenzo's whole empire because the U.S. Pension Benefit Guaranty Corp. wasn't letting him walk way from it. So Lorenzo's empire, which included Continental Airlines and Continental Airlines Holdings (nee Texas Air Corp.) went under, as I predicted. But Lorenzo managed to bail out ahead of the crash, which I hadn't predicted.

Once you realize that junk bonds were gone, you realized that junk bank loans would vanish too. Which is why I was sure that the two employee bids for United Airlines would fail, as they did.

When it came to writing about Macy's, though, I got too cocky. In September, I wrote that the company's claim that it would sell new stock to investors at five times the price investors paid in the 1986 Macy's leveraged buyout was absurd. And it was. But instead of stopping while I was ahead, I implied in early December that Macy's wouldn't be able to sell $100 million of new stock at the 1986 price. Oops. In mid-December, Macy's proved me wrong by selling more than $150 million of stock. I'm still pessimistic about the company, but I'll be more careful next time.

When I stuck to the obvious, I did better. I wrote in June that a used-car company called Urcarco, which had just sold more than $100 million of new stock to investors for $19.875 a share, was making money on its bookkeeping, not its operations. The stock recently traded at less than $4. I also wrote that Kohlberg, Kravis Roberts & Co. had to do something to salvage the RJR Nabisco leveraged buyout -- which it did by putting up $1.7 billion of new cash.

In June, I wrote that Walt Disney Co.'s issue of Liquid Yield Option Notes was goofy, but that investors would buy tons of the stuff, even though it was awful. I was right. My article, which my children objected to, saying that I was showing disrespect for Bambi, prompted the following macho missive from Disney Treasurer Richard Nanula after the LYONs were safely sold: "In your face." What do you expect from someone with a mouse on his letterhead?

I haven't been as clear as I should have been about First Executive, the Los Angeles-based holding company for the junk-laden Executive Life insurance companies in New York and California. I owe all you Executive Life policyholders an explanation of how I can say that First Exec is busto but the Executive Lifes may be sound.

And here it is. First Exec has defaulted on its bank loans and preferred stock. But the hundreds of millions of dollars First Exec raised with the borrowings and preferred stock sales were invested in the insurance companies as capital. The insurers don't have to repay it -- that's First Exec's problem. So the insurance companies have more capital than First Exec, and fewer liabilities. That's why they may still be sound.

The year wouldn't be complete without Meshulam Riklis, whose deals are so complicated that I get a headache just thinking about them. I'm not sure whether I was right in writing that his McCrory chain was bankruptcy bait. I'm still trying to figure out what's going on, and someday, hopefully in 1991, I'll let you know.

Allan Sloan is a columnist for Newsday in New York.