Ever since late 2009, when it began the rounds of Treasury and mortgage-backed security purchases that eventually caused its balance sheet to balloon from less than $1 trillion to about $4.5 trillion, the Fed has been promising to slim down again. The faster it does so, the better, because that fat balance sheet means the credit that could be supporting more robust economic growth is instead being shunted into the U.S. Treasury and the housing market.

Yet despite hopes that Wednesday’s FOMC minutes would at last divulge a definite plan for shrinking the Fed’s balance sheet, those minutes added little to the now-standard line that the Fed wouldn’t start shrinking “until normalization of the level of the federal funds rate was well under way.”

Alas, that standard line is all one needs to worry that the Fed’s plan may not work. Why is that so? Because making further rate hikes a condition for finally going on a diet could mean having to choose between putting that diet off indefinitely, or tightening credit to a point that risks triggering another recession.

That is the case because shrinking the Fed’s balance sheet itself tightens credit by reducing the cash reserves available to banks and other financial institutions. Fewer reserves mean fewer bank deposits and loans. Indeed, since quantitative easing caused the Fed’s balance sheet to grow in the first place, one might even call the Fed’s plan to shrink again “quantitative tightening.”

Yet, the Fed’s plan also calls for raising the federal funds rate — the Fed’s primary tool to manipulate interest rates. These days, that means paying banks a higher return on their voluntarily held or “excess” cash reserves, while also raising the (lower) rate that the Fed pays to borrow cash from other financial institutions through its “reverse repo” program.

The actual fed funds rate bounces within a range between these policy rates — presently from 0.75 to 1 percent. Raising that range enhances financial institutions’ appetite for cash, encouraging them to hoard it instead of trading it for other assets. So rate hikes also tighten credit.

In short, the Fed’s normalization plan calls for it to prop-up banks’ demand for cash, as a prelude to reducing the supply of cash! That means tightening and more tightening. The rub, of course, is that conditions may never justify so much tightening. What’s more, no plan for Fed normalization can work that would prevent the Fed from meeting its overarching inflation and employment targets.

Is there less fraught path back to normal? There is, and it’s one the Fed can start down without delay. Instead of making higher rates a condition for shrinking its balance sheet, the Fed should make an unconditional commitment to begin shrinking its balance sheet according to a definite schedule. It can then adjust its policy rates as needed to achieve its broad policy goals.

That could well mean having to gradually reduce those rates as cash reserves decline to keep money from becoming too tight. The idea is to make excess reserves less desirable as they become less available.

At some point, banks will revert to their old practice of holding only minimal reserves and actively lending and borrowing on the fed funds market to dispose of unwanted reserves or to make up for temporary shortages. Once that happens, the Fed can revive its pre-crisis practice of setting a specific market-determined fed funds rate target, which means that it can also cease altogether to pay interest on banks’ excess reserves.

It can, at the same time, shut down its reverse-repo program. In other words, the Fed, besides shrinking its balance sheet, will have successfully restored its pre-crisis methods of monetary control, which is what policy “normalization” ought to mean.

The presence of a more certain path back to a leaner, meaner Fed doesn’t mean that the way is free of obstacles. Keeping monetary policy on track while reversing the massive changes that have taken place since 2008 is bound to be challenging. Still that reversal is badly needed; and a merely difficult plan for achieving it is a heck of a lot better than one that might never get us there at all.