Over the past year, it's been Us vs. Goldman Sachs. Conspiracy theories have multiplied aplenty. Loathing hasn't been shy. Heads have been called for. That much is certain.

But what would a sensible Goldman Sachs (NYSE:GS) look like? One that we could put up with, without being the butt of every egg-throwing anti-Wall Street protest imaginable? To answer that question, we might want to go back and look at the Goldman Sachs of years past, when it commanded respect.

Goldman Sachs went public in 1999. Before that, it was a private partnership, structured the same way many law firms and accounting firms are. Partners were senior employees who, after years of devastatingly long hours, were selected into a small fraternity of owners.

These partners owned the entire firm. All of the capital was theirs. They literally supplied it out of their own earned income. As Charles Ellis writes in his book, The Partnership: "Goldman Sachs retained most of each partner's yearly earned income. As a result, anyone who became a partner in Goldman Sachs usually experienced a drop in spendable income."

Being directly supplied by Goldman's partners, capital was watched after like a hawk. It was not only their livelihood, but also a liability that, if abused, could result in personal failure. Lisa Endlich's 1999 book, The Culture of Success, describes just how important an issue prudence was:

In a private partnership none of the assets of partners are shielded from liability, and the individual partners are exposed down to the pennies in their children's piggy banks ... The actions of a rogue trader could spell personal bankruptcy ...

This created a culture of discretion. When a trader down the hall could throw you into personal bankruptcy, partners were understandably berserk about risk management. Employees were essentially conducting business out of their boss's personal checking accounts. Accordingly, business was conservatively managed for the long term.

Just how long-term? At one point, retiring partners could only sell 25% to 50% of their partnership interest inside the first year of leaving the company. The rest could be sold over a five-year stretch. In 1996, amid heavy losses, management prolonged that period in order to maintain capital. Retired partners had to cover Goldman's losses with their personal wealth. That's commitment.

During the mid '90s, most major investment banks had gone public, except for Goldman Sachs. Maintaining its partnership culture fostered stability. As the New York Times wrote in 1998, "One Goldman banker described this as 'getting rich slowly.' There are stars in the Goldman system. But they get smaller up-front rewards, and less public recognition, than they might get elsewhere."

After going public in 1999, those rewards began to shift. Suddenly, employees weren't playing entirely with partners' money. They were playing with shareholders' money. Partners were far less jeopardized by screwups. The shareholders supplied the capital. The shareholders bore the risk. Employees were gambling with someone else's money, and for the most part, they didn't know who that someone was.

The biggest change coming from this was shifting focus from investment banking into risk-taking with the firm's own capital.

In 1998, before Goldman went public, no-frills investment banking and asset management made up 72% of revenue. Investment banking is boring. You help AT&T (NYSE:T) raise capital. You advise Microsoft (NASDAQ:MSFT) on acquisitions. You take the next Google (NASDAQ:GOOG) public. It's a business built on expertise and hard work. It's safe. It's mostly riskless. Like a law firm.

But by the third quarter of 2009, 81% of revenue came from trading and principal investments. Boring investment banking made up just 7% of revenue.

During Goldman's partnership days, you would never see risk-taking of this magnitude. The partners wouldn't have been able to sleep at night. It's too much dynamite to carry home every day. One slight hiccup, and not only would a partner's Goldman stake be wiped out, but also their kid's college education, their watches, yachts, cars, mansions, and jets. They were personally liable for the firm's exposure.

Why is this important? Because the compensation system on Wall Street is still a game of "heads I win, tails someone else loses." You can't blame Wall Streeters for exploiting this. They're acting rationally. But the system is conducive to devastating, economy-wrecking speculation.

Warren Buffett said it best:

What you have to change in Wall Street is you have to make sure that in addition to carrots, there are sticks. And it can't be a one‑way street where they are making ungodly amounts of money when things are good and then they move on to someplace else for a while when things are bad. You have to create a downside ... You have to put in something where there is downside to people who really mess up large institutions ... Too many people have walked away from the troubles they have created for society, not just for their own institution, and they have walked away rich.

This isn't just about Goldman. Bankers at JPMorgan Chase (NYSE:JPM), Citigroup (NYSE:C), and Bank of America (NYSE:BAC) all enjoy a similar setup.

And this doesn't just apply to banks that were bailed out: Plenty of lucky souls at Washington Mutual and Lehman Brothers walked away with stupendous net worths thanks to the reckless risk-taking of prior years.

The Goldman partners of yesteryear didn't have less ambition. They didn't have less intelligence. They didn't even have less desire to become rich. What they had were different incentives. And if we're going to fix our financial system, focusing on those incentives should be near the top of the list.

Fool contributor Morgan Housel doesn't own shares of any of the companies mentioned in this article. Google is a Motley Fool Rule Breakers selection. Microsoft is a Motley Fool Inside Value selection. Motley Fool Options recommended a diagonal call on Microsoft. The Fool has a disclosure policy.