Dennis Lockhart just entered his ninth year as president of the Federal Reserve Bank of Atlanta, and during that time he has never voted to raise interest rates. He took the job during the very early days of the financial crisis, and he has never really had the chance. But that is about to change. Mr. Lockhart said in an interview Monday that he expected he would vote to start raising rates by September at the latest.

He also talked about his optimism that the economy was gaining strength despite signs of a slow start to 2015, and he dismissed a proposal by Richard Fisher, the recently retired Dallas Fed president, to reduce the role of the New York Fed in making monetary policy.

A lightly edited transcript of our conversation follows:

Q. You predicted that economic growth would accelerate in 2015, but there are signs that this year, like last year, is off to a slow start. Have you reconsidered?

A. The slowness in the first quarter obviously raises concerns that we're going to see a continuing or persistent slowdown, but that's not my base case view. My base case view is that we'll see a rebound in the second and third quarter and beyond and that we'll stay on the basic track that has been our story, our narrative here, for the last year or more. And that is a 2.5 percent to 3 percent growth rate with continuing improvement on the employment front, and gradual rise in inflation toward the 2 percent target. So to some extent I'm taking on a Wilbur Mills position: That's my story and I'm sticking to it.

Q. There is always some uncertainty. Does the current confusion seem any greater than normal?

A. It does not feel like it's extraordinary. What makes it a little different set of circumstances is the prospect of making a very serious policy decision in the coming months. When you combine the decision that I genuinely believe that we will make with the ambiguity, it is not as comfortable a decision-making environment as one would prefer. Maybe you never have a totally comfortable decision-making environment. I'm sure that's the case. We've had ambiguity in periods before — quite a bit of it over the last several years — but we were also not facing what I would call a historic decision to go from one era to another era.

Q. I'm surprised to hear you describe the first rate hike as a historic moment, because you've argued that people are overly focused on it.

A. Too much can be made of the liftoff decision itself because the really important factor is what's going to be the interest rate environment for the foreseeable future and to what extent is it going to remain accommodative, which I believe it will. That's more important to the real economy than the exact date of liftoff. But nonetheless it's going to be treated as a significant decision and a significant phasing from one period to another. Certainly I think when the historians write about it, it will delineate.

Q. Of course, like most of your colleagues, you've never actually voted to raise interest rates.

A. No. I've voted to reduce them but never to raise them. Actually, [former Philadelphia Fed president] Charlie Plosser, on his retirement earlier this month, we pointed out that he would be the longest-standing president never to have raised interest rates. And we asked him, "Is that really the way you wanted to be remembered?"

Q. You also gave Atlanta Hawks hats to Mr. Plosser and outgoing Dallas Fed president Richard Fisher. Did you get one for yourself?

A. I didn't have one made for myself. I haven't found any sports team called "Centrists" or "Moderates."

Q. You've said that you expect the Fed to start raising rates at one of three meetings: June, July or September. Are you confident it will happen by September?

A. I wouldn't say I'm 100 percent confident. I noted that Stan Fischer said, "This year." So for me to say June-July-September full confidence is probably overstating it, but I think it's quite likely. And if we were to go beyond September, it would be because we were really disappointed in the stream of data that come in.

Q. O.K. Does that mean are you confident it will happen by the end of the year?

A. I think it's highly likely this year. If we reverted to a decision point after the end of the year, it would probably reflect either a shock to the economy that really changed the trajectory of the economy or we would have been misreading something pretty seriously.

Q. So you're comfortable that labor market slack is being reduced to the point where it's appropriate to raise interest rates?

A. I'm comfortable with the notion that we can make a decision in the coming meetings to begin a process that will be itself gradual with some slack remaining in labor markets. Probably there is some amount of slack left that remains to be absorbed. But having said that I'm comfortable that we are closing in on something that would be close to full employment conditions and that makes this kind of decision appropriate.

Q. The Fed historically has erred on the side of suppressing inflation, even at the expense of tolerating higher unemployment. If you raise interest rates in the next few months, do you risk perpetuating that pattern?

A. I do not see that as a good description of current circumstances. The way I would characterize this period going back a few years in the recovery period is that there is never going to be complacency about inflation. Central bankers are not built that way. Should not be built that way. But if anything there was a bias toward looking after the employment side of the mandate.

I am not one who's too concerned at the moment with the risk of cratering the recovery by acting too early. Nor am I concerned that we are going to see a breakout in inflation by delaying that decision if there were other reasons to do so. I don't see the risks as high in either direction. If we make a decision early — early being June — or if we make a decision later, September or even later, in the greater scheme of things I don't think it's going to matter dramatically. I think the economy can absorb that almost on any of those dates. I'm just not highly concerned that there is a risk of a policy error based upon the difference of one or two meetings.

Q. The Fed has been pretty clear in describing lower oil prices as good for the economy, but some economists see evidence that at least in the short term, the costs may be piling up faster than the benefits. What is your view about the economic impact?

A. I don't have an extremely well-studied opinion on that. We've seen some effects in investment and we're beginning to see some effects on employment in the oil and gas sector, and the benefits to the consumer that were expected to come from more discretionary spending power have not materialized in any dramatic way yet. So what we have is some clear effect in the short term in the oil and gas sector and not yet seeing the benefit develop. The savings rate has gone up a point, and there seems to be a reasonable argument that consumers are not yet prepared to change their spending patterns because of this, maybe expecting that perhaps prices could rise again.

Q. The Fed has made a real point of playing down the recent decline in market-based measures of inflation expectation, suggesting that survey-based measures are more accurate. This is not a distinction the Fed has made historically. So why now?

A. My view is that the whole tool of following inflation expectations has certain — that one should not rely literally on either tool as really nailing down inflation expectations. Surveys of consumers at least in the short term are highly influenced by prices at the pump. So as gasoline prices go, so go many consumer views as to the inflationary environment. And there are a number of market influences on inflation compensation, so translating from inflation compensation to inflation expectations is not necessarily a straightforward calculation. So I think we have to use every tool we have to form a judgment as to the key question of whether the broad consuming public — and for that matter businesses that are spending — are changing their view in a fundamental way of the future path of inflation. I do accept the view that even though inflation expectations are hard to truly get a handle on, they are influential in the behavior of consumers and businesses.

Q. And are you confident that we have not seen such a shift?

A. Pretty confident.

Q. What's so bad about 1.5 percent inflation? A lot of people struggle to understand why the Fed would ever try to drive up inflation.

A. We want inflation comfortably away from the threshold of deflation. Two percent, there is an element of that being an arbitrary number, but it is a number that is far enough away from the threshold of deflation to provide a cushion that covers both distance from deflation and measurement error. And at the same time 2 percent is low enough that it should not distort decisions that households make or businesses make for their long-term interest. So it's a happy medium that I think is a reasonable target. If we were chronically below that but static around a number, like we are today, like 1.3 percent or 1.4 percent, I'm not sure it's going to do substantial harm, but it's reflective of an economy that's weaker than the ideal. If we see that number beginning to decline, meaning headed toward zero, then I think you have a serious matter on your hands that monetary policy is to some extent capable of dealing with.

Q. Do you agree with Richard Fisher, the recently departed Dallas Fed president, who argued in one of his last speeches that the New York Fed should be stripped of its outsize role in making monetary policy? (The head of the New York Fed is the only reserve bank president who has a permanent vote on the Federal Open Market Committee — the other 11 banks rotate through four voting seats.)

A. I don't see exactly what the problem is. People ask me all the time, "What's it feel like to be a voter?" I don't see quite frankly that it's a big important role in the greater scheme of things. What really counts is your voice at the table. And a nonvoting president with an extremely cogent argument or a particular way of framing an important question that we're facing would have far more influence at the table than a voting president who maybe is not that inspired at that particular meeting. So I just don't make too much of the voting power when I have it or when I don't have it. What really counts is the dynamic that takes place around the table in coming to a consensus. I don't want to say that the votes are perfunctory or that they would never matter, but I've been in the Fed now for eight years and they have not been close calls. By the time the committee comes around to voting, it's pretty clear what the working consensus is.