聯準會理事 Stephen Miran 深談總經、AI、穩定幣與美國金融創新

0:00聯準會的雙重使命與近期利率決策

0:00Miran 理事，很高興你來到這次數位資產高峰會。歡迎。謝謝邀請我來。我先澄清一下，我是七位聯準會理事之一，不像英格蘭銀行那種總裁。聯準會有七位理事。Governor Miran, great to have you at the Digital Summit. Welcome. Thanks for having me. And let me just clarify that I'm one member of a board of seven governors. There are seven of us. I'm not a governor in the sense of governor of the Bank of England or something.

0:15Miran 理事，我想先從上週最新一次 FOMC 會議談起。你在那次會議中支持降息 25 個基點，這是個不同意見。能不能談談你目前的決策框架，為什麼你會這樣表態？Governor Miran, I would love to just start by unpacking the most recent FOMC meeting that occurred last week. You dissented in support of a 25 basis point rate cut, so I'd just love for you to unpack your current framework and why you dissented there.

0:31好的，謝謝你。你知道，在聯邦準備理事會（Federal Reserve），我們有雙重任務。我們受到國會的指派，要關注通膨、維持物價穩定，以及勞動市場。我的看法是，儘管實際測得的通膨很高，但在這個任務的通膨部分，其實並沒有那麼大的問題，因為超出目標的通膨大多是由於某些衡量方式上的特殊因素，例如投資組合管理服務，這基本上就是股市上漲，導致我們衡量通膨的方式被高估了30到40個基點。Sure, thanks. Look, you know, at the Federal Reserve, we have a dual mandate. We are tasked by Congress with paying attention to inflation, stable prices, and to the labor market. My view is that the inflation side of the mandate has, despite high measured inflation, has not been so problematic because a lot of the inflation excess over target has been a result of some quirks of measurement of things like portfolio management services, which is basically just the stock market going up, biasing the way we measure inflation up by 30 to 40 basis points.

1:01一旦你把那些因素調整過後，我並不認為通膨是特別嚴重的問題，反而勞動市場在過去三年中一直呈現非常緩慢的走弱趨勢。Once you correct for those things, I do not view inflation as being overly problematic, whereas the labor market has been on a very gradual trend of weakening over the course of the last three years.

1:10這個趨勢一直存在，而且持續進行中。你可以從新進入勞動市場的人越來越難找到工作看出來。你也可以從人們在失業期間換工作的時間越來越長看出來。因此，我的看法是，經濟還能承受來自貨幣政策對勞動市場的額外支持。當然，最近幾週發生的很多事情，都和伊朗戰爭有關。我認為，油價的變動讓很多人感到恐慌。你知道，從交易的角度、從市場的角度來看，行情會在兩個方向劇烈波動。That trend has been in place; it's been continual. You see it in increasing difficulty of finding jobs for new entrants to the labor market. You see it in increasing length of time people spend between jobs during unemployment. And so my view is that the economy can bear additional support for the labor market from monetary policy. Now, a lot of what's been going on in the last few weeks has, of course, been tied to the war in Iran. And I think what's going on in oil prices has spooked a lot of people. You know, and I understand from a trading perspective, from a markets perspective, things whip around wildly in both directions.

1:45你會看到新聞標題來回跳動，這會影響市場。如果你有槓桿部位，那種感覺會非常、非常地強烈。但從貨幣政策的角度來看，我們必須為未來12到18個月制定政策，因為貨幣政策對經濟產生影響有很大的時滯。如果我們調整利率，實際傳導到經濟成長、實際失業率和通膨，至少要一年。所以我們需要為一年到一年半之後設定政策，而油價變動對經濟的影響其實比這個時程快得多。You get headline ping pong, and that pushes markets. And if you have levered positions, that can feel very, very, you know, sort of very intense. But from a monetary policy perspective, we have to make policy for 12 to 18 months from now because there are big lags with which monetary policy hits the economy. If we adjust interest rates, it doesn't feed through into actual economic growth, into actual unemployment and inflation for at least a year. So we need to set policy for a year to a year and a half out, and the way that changes in oil prices affect the economy is actually much faster than that.

2:18當你看到油價飆升，當你因為像伊朗發生的事情導致油價上漲時，油價會立即上升，而短期內整體通膨也會大幅上升。但如果你把時間拉長到一年到一年半之後，很不可能這會造成後續影響，進而在一年到一年半後還影響經濟，而這正是貨幣政策可以發揮作用的時點。因此，這也是為什麼傳統上中央銀行應該忽略油價衝擊的經典理由之一，就像聯邦準備理事會（Federal Reserve）過去一向的做法一樣。When you get a spike up in oil prices, when you get a move higher in oil from something like what's going on in Iran, the oil price goes up immediately and headline inflation goes up a lot in the short term. But as you look a year to a year and a half out, it's very unlikely that that causes subsequent effects that are affecting the economy a year to a year and a half out, which is when monetary policy can affect the economy. And so that's part of why the classic reasoning of why a central bank should look through an oil shock, as the Federal Reserve has historically done.

2:55供給衝擊、AI 與監管鬆綁的影響

2:55看起來每隔幾年我們就會遇到某種衝擊需要去應對，我很好奇你能不能幫我們拆解一下這個現象。討論的焦點似乎在於，很多這些供給衝擊似乎都朝著同一個方向，也就是導致通膨上升。你怎麼看這個問題？是的，我想說幾點。首先，當你思考這些負面的供給衝擊如何影響經濟時，像是油價上漲這類情況，就像我之前說的，它在短期內會非常明顯地影響整體通膨，但如果你把時間拉長來看，其實對其他方面沒有什麼影響。It seems like every few years we have some sort of shock that we have to navigate through, and I'm curious if you could just unpack it, you know. The discussion seems to be that a lot of these supply shocks seem to go in one direction, one being higher inflation. How do you think about that? Yeah, so I'll say a couple things. One is when you think about how these negative supply shocks are hitting the economy, things like oil prices moving higher, as I said before, it affects headline inflation very much in the short term, but as you look further out, it doesn't affect anything.

3:27也就是說，如果你開始擔心這些負面的供給衝擊，那你應該要觀察的就是未來一段時間的通膨預期是否上升。如果你去看通膨交換市場，你會發現一年後、兩年後、三年後的遠期通膨預期，其實幾乎都沒有受到最近發生的事情影響。事實上，很多數據甚至比——抱歉，很多數據甚至比之前還要低。自從我們一月見面以來，自從FOMC一月開會以來，這些數據都變得更低了。And that is, if you get concerned about these negative supply shocks, that's what you would be looking for. You would be looking for a rise in inflation expectations further out in time. And if you look at the inflation swap market, you see that forward inflation expectations a year out, two years out, three years out are all pretty much unaffected by what's been going on. And in fact, a lot of them are lower since the start of the— sorry, a lot of them are lower. Since we met in January, since the FOMC met in January.

3:53因此，石油價格負面供給衝擊對於較長期的通膨預期，完全沒有任何滲透現象，也就是說，這些影響並沒有延伸到未來，而這正是貨幣政策能夠發揮作用的時間點。And so there's been zero bleed-through of the negative supply shock of oil prices into inflation expectations further out, you know, further out in time, which again is when monetary policy will be able to affect.

4:06你還會擔心的另一件事，就是如果出現工資-物價螺旋。當物價上漲，然後工資也跟著上漲，這又進一步推高物價，於是你就會產生一個負面的反饋循環，這會造成一個自我實現的通膨螺旋，必須由貨幣當局透過緊縮政策來抵消並壓制。這並不是我們現在看到的情況，因為就像我之前說的，勞動市場大約三年來一直處於非常緩慢的降溫趨勢。隨著勞動市場非常緩慢地降溫，以及工資壓力逐漸減弱，你就不會出現工資-物價螺旋。The other thing you'd be concerned about. As if there were a wage-price spiral. If prices go higher and then wages go higher, and that pushes prices higher again, you get a negative feedback loop that just creates a self-fulfilling inflationary spiral that requires tightening from the monetary authority to offset and to squash. That's not something that we see happening right now, because the labor market, as I said before, has been on this very gradual cooling trend for about three years now. And with the labor market very gradually cooling and declining wage pressures, you just are not going to get a wage price spiral.

4:43這通常會是中央銀行會做出回應的那種類型的事情。現在，你剛才還提到另一件我想要延伸討論的事。那就是我們過去經歷了很多負面的供給衝擊，這種情況一直以來都有。我認為人們往往會認為，這些就是我們所經歷過的唯一供給衝擊。但事實並非如此。其實也有正面的供給衝擊正在影響經濟——這些也同樣受到很多關注。其中之一就是 AI，對吧？AI 就是一種正面的供給衝擊。它提升了經濟的生產力，讓人們能用更少的生產投入產出更多。That would be the type of thing that a central bank would normally respond to. Now, you said something else that I'd like to pick up on. Which is that we've had a lot of negative supply shocks, you know, sort of over time. And I think that there's a tendency to think that those are the only supply shocks that we've experienced. And that's not the case. There are also positive supply shocks that are hitting the economy too, and— These get a lot of attention too. So one of them is AI, right? AI is a positive supply shock. It increases the productive capacity of the economy. It lets people produce more with less inputs into production.

5:19那是一個正向的供給衝擊。另一個非常強大的正向供給衝擊，是來自於去管制的趨勢。現在，我認為這裡是一群專注於加密貨幣的人。法規很可能在你們許多人的商業模式中扮演了重要角色，而必須應對這些法規，對在場許多人來說，在某些時刻，可能讓你們的進度變慢、減少你們能夠產出的東西、提高生產成本。而隨著這些法規障礙逐漸消退，你們就更容易製作出人們想要購買的產品。因此，當法規環境變得更寬鬆時，這也是一個對經濟產生影響的正向供給衝擊。That's a positive supply shock. Another very powerful positive supply shock that's been hitting the economy has been the trend in deregulation. Now, I think this is a group of people who are focused on crypto. Regulations have probably played a large part in a number of your business models, and having to deal with those regulations has probably, for a number of people in this room at some point, slowed you down, reduced the amount of things you can produce, increased the cost of production. And as those regulatory barriers recede, it becomes easier for you to make products that people want to buy. And so as the regulatory backdrop becomes easier, that's a positive supply shock that hits the economy too.

5:56我在一月份於希臘發表了一場演講，當時我檢視了現代有關監管的文獻。要量化聯邦監管法規並不容易，並且要以經濟學家希望進行嚴謹實證研究的方式將其轉換成一個數字更是困難。不過，我參考了現代文獻中運用 AI 和機器學習工具來進行這類工作的研究，並計算出過去大約十五個月來席捲經濟體的去監管浪潮，最終每年大約會讓通膨減少半個百分點。在接下來的幾年裡，對吧？所以，這其實是相當可觀的。Now, I gave a speech in January in Greece where I looked at the literature, the modern literature on regulation, and it's very difficult to sort of quantify the federal regulatory code. And turn that into a number in a way that economists would like to do rigorous empirical studies. But I looked at the modern literature that sort of uses AI and machine learning tools to do so, and I calculated that the deregulatory wave that's been hitting the economy over the last 15 months or so would ultimately drag on inflation by about half a percent a year. For the next few years, right? So that's, that's, that's, that's quite chunky.

6:37就在幾週前，聯邦準備理事會有兩位經濟學家，Danilo Cascaldi-Garcia 和 Matteo Iacoviello，發表了一篇新的聯準會研究報告。這份關於放寬監管的聯準會研究報告發現，他們採用了與過去文獻完全不同的估算方法，也用了與過去文獻完全不同的監管衡量方式。而當你將他們的研究結果應用到過去大約十五個月我們所看到的放寬監管衝擊範圍時，這意味著對通膨會有大約 0.3% 的拖累。這個影響會在未來兩年每年都出現，對吧？所以這不只是一個一次性的效果，而是來自監管環境逐漸減弱所產生的持續性通膨下行效應。Now, just a couple of weeks ago, there was a new Federal Reserve staff research paper released by two staff economists at the Fed, Danilo Cascaldi-Garcia and Matteo Iacoviello, and what this Fed research paper on deregulation found is they did an entirely different estimation method than the previous literature, and they did an entirely different measurement method for regulations than the previous literature. And when you apply their results to the scope of the deregulatory shock that we've seen over the last 15 months or so, it implies a roughly 0.3% drag on inflation. Each year for the next two years, right? So that's not just a one-off effect. That's a persistent disinflationary effect from the receding regulatory backdrop.

7:22現在，使用完全不同的文獻、完全不同的估算方法，我算出來是0.5。他們算出來是0.3。我認為這兩個數字大概都在誤差範圍內。它們當然都在信賴區間之內。我沒辦法拒絕其中一個而支持另一個。但不論哪一個，這個數字都很大，而且我認為非常重要。我認為這屬於決策者在思考正確貨幣政策設定時應該納入考量的衝擊類型。所以，沒錯，確實有許多負面的供給衝擊影響到經濟，像是油價上漲這類的事情。Now, using an entirely different literature, entirely different estimation methods, I calculated 0.5. They calculated 0.3. I think those are probably within noise of each other. They're certainly within confidence bands. I wouldn't be able to reject one in favor of the other. But either of them are very big, and I think very substantial. And I think of the types of shocks that policymakers ought to take into account when we're thinking about the correct settings for monetary policy. So yes, it is the case that we have had a number of negative supply shocks hit the economy, things like oil moving higher.

7:53但同時，我們也看到非常強大的正向供給衝擊正在影響經濟。而且，與石油不同的是，我預期其中一些因素，例如放寬管制和人工智慧，將會持續帶來通膨壓力下降的效果。But it's also the case that we have very powerful positive supply shocks hitting the economy too. And unlike oil, I expect some of these, like deregulation and AI, to be persistent in their disinflationary effects.

8:04政策展望：利率、正負供給衝擊與中性利率

8:04所以上週的會議是一個經濟預測摘要更新會議。我很想聽聽看，當你在平衡這些正面或負面的供給衝擊時，這會如何影響你對明年聯邦基金利率的預測？在上一次的預測摘要中，因為油價衝擊，對吧，那會非常快速地影響經濟，而我認為其他一些衝擊則是隨著時間慢慢發酵。因此，我確實將我今年的總體通膨預測上調到2.7%，對吧？所以因為油價衝擊，我把預測稍微調高了一點。So last week's meeting was a summary of economic projections update meeting. I would love to just hear about when you balance out the context of either these positive or negative supply shocks, how does that characterize your forecast for the federal funds rate for the next year? So, in the last summary of projections, because of the oil shock, right, that hits the economy very quickly, whereas I think some of these other shocks play out over time. I did raise my inflation projection for my headline inflation projection for this year to 2.7%, right? So I moved it a little bit higher because of the oil shock.

8:42然而，正如我之前所說，這並不是政策應該回應的類型，因為這些影響都是一開始就發生，而政策對經濟的影響則是在12到18個月之後。現在油價上漲，對12到18個月後的通膨幾乎沒有什麼影響。所有的通膨效應都是一開始就發生的。12到18個月後真正會發生的，是經濟可能會變得比較疲弱，因為每一塊人們花在加油的錢，就是沒有花在其他商品和服務上的錢。因此，這可能會對失業率產生上行壓力。所以。However, as I said before, this is not the type of thing that policy should respond to because that happens all up front, and policy affects the economy 12 to 18 months out. Oil moving higher now has very little inflationary consequence 12 to 18 months out. All the inflation happens up front. What does happen 12 to 18 months out is the economy might be weaker because every dollar that people put into, you know, filling their gas tanks is a dollar they're not spending on other goods and services. And so that might put upward pressure on the unemployment rate. So.

9:12在三月的經濟預測摘要中，我把我的政策利率提高了半個百分點，這不是因為伊朗的石油，而是因為我們在十二月的經濟預測摘要和三月的經濟預測摘要之間收到的通膨數據。這讓我的預測大約落在中性水準，對吧？所謂中性政策利率，就是既不具刺激性也不具寬鬆性的貨幣政策利率。現在我認為這大概是在百分之二點五到百分之二點七五左右。我們目前大約比這個水準高出七十五個基點。所以，你知道，我認為——抱歉，我們，對，我們現在大約比那個水準高出一個百分點。In the March Summary of Economic Projections, I boosted my policy rate by half a percent, not due to oil in Iran, but due to the inflation data that we received in between the December Summary Projections and the March Summary Projections. That puts my projection at about neutral, right? So the neutral policy rate is the monetary policy interest rate that's neither stimulative nor accommodative. Right now, I think that's probably about two and a half percent, two and three-quarters percent. We're about 75 basis points above that level. So, you know, I think it's—sorry, we're, yeah, we're about a percentage point above that level right now.

9:55我認為在今年內，讓政策回到中性是適當的。我不認為經濟需要像2021年或2022年那樣，靠貨幣政策大力踩油門來加速經濟。但我也不認為經濟需要被壓抑。而現在的政策是略為緊縮的，確實在拖慢經濟。我認為這和目前的總體經濟背景並不一致。I think it's appropriate over the course of this year to just get back to neutral. I don't think the economy needs monetary policy to be slamming on the gas and accelerating the economy like it was in 2021 or 2022. But I also don't think it needs to be holding the economy back. And right now it is modestly restrictive, and it is holding the economy back. And I don't think that's consistent with the macroeconomic backdrop.

10:19你剛剛一開始提到這個 AI 潛在生產力爆發的想法，還有它可能比像負面的石油供應衝擊這種事件更持久。而我第一個想到的，就是這種持久性的衝擊，對長期經濟中性的利率，或所謂的 R-star，可能會有什麼影響？我很好奇你怎麼看待這種潛在強大、正面的衝擊，像 AI 生產力爆發，對於中性利率長期預測的影響。對，所以我的看法是，我認為 AI 生產力爆發毫無疑問會把中性利率推得更高。So early in there, you mentioned this idea of an AI potential productivity boom and how it could be much more durable than something like a negative oil supply shock. And the first thing that comes to my mind there when I think about a durable—Shock like that is, what is its potential impact on the long-term economic neutral rate, or R-star? I'd be curious to hear about how—what's your perspective on the more long-term forecast of the neutral rate in light of such a potentially powerful shock, positive shock, such as the AI productivity boom. Yeah, so look, you know, I think that the AI productivity boom does unambiguously push the neutral rate higher.

10:57然而，我認為還有許多其他相互抵銷的因素一直在影響中性利率。因此，確實如此，如果投資隨著時間變得更有利可圖，如果投資隨著時間變得更有生產力，那麼這會提高長期的中性利率。你知道，就是你在資本投資上獲得的中性利率，這會推升中性利率。但還有其他因素一直在壓抑它。而我一直試圖引起大家注意的一個對中性利率有強大影響的因素，就是人口成長率的變化。我們經歷了人口成長率在兩個方向上最大的衝擊。However, I think there's a lot of other countervailing factors that have been weighing on the neutral rate. And so it is the case that if investing becomes more profitable over time, if investing becomes more productive over time, that raises the long-term neutral. You know, neutral rate of interest that you get on capital on investing, and that boosts neutral. But there's other things that have been weighing on it. And one thing that I've been trying to draw attention to that's been weighing on it, I think quite powerfully, is the change in population growth. We lived through the biggest shocks to the population growth rate in both directions.

11:35在我這一生當中，甚至可能在在場許多人的一生當中，在短短幾年內，我們經歷了一段人口大幅成長的時期，大約是在2021年到2023年左右。然後到了2024年、2025年、2026年，這個人口成長幾乎急遽下滑，現在勞動年齡人口的成長大概已經接近持平了。所以這是一個成長大幅上升的高峰，接著又大幅下跌。就像中性利率是資本均衡報酬率的函數，而這又受到生產力的影響，同時也受到經濟成長率的影響，而經濟成長率又受到人口成長的影響。In certainly in my lifetime, probably in the lifetimes of many, many folks in this room, within the course of a few years, we had a period of massively growing population in 2021 to 2023 or so. And then in 2024 to 5, 6, that population growth plummeted pretty much to, you know, working age population growth is probably, you know, pretty close to flat now. And so that was a huge spike up in growth and then a huge spike down. And just as the neutral interest rate is a function of the equilibrium returns in capital that are affected by productivity, it's also affected by the growth rate in the economy, which is affected by population growth.

12:16所以在我看來，這是一個對利率產生非常重大影響的因素。你可以在全世界各地看到這種現象。在疫情爆發之前，我們花了很多年討論全球利率的日本化。你知道，生育率下降、全球人口老化、人口成長率下降，這些都導致許多國家的利率走低。當時有很多時間都在討論，大家是不是都會走向這個方向，對吧？我認為這些管道、這些路徑一直都是有效的經濟路徑，也許在疫情之前它們是最主要的重要因素，當時有很多人在討論這些問題。And so that, in my view, is a very big thing weighing on interest rates. And you see that across the world. We spent many years before the pandemic discussing Japanization of global interest rates. You know, declining fertility rates, declining, you know, aging populations around the world, declining population growth led to lower interest rates in a lot of countries. And there was a lot of time spent discussing, is everybody going in this direction, right? I think that those channels, those pathways were always valid economic pathways, and maybe they were of primary importance sort of before the pandemic, and a lot of people were talking about them.

12:55然後在疫情期間，很多其他的事情開始變得重要，主要是跟疫情有關，還有隨之而來的大規模經濟支持計畫。但那些——人口老化、人口成長和利率之間的關聯並沒有消失，對吧？只是，有其他事情更吸引大家的注意力。但那些關聯路徑依然有效，我認為在接下來的幾年裡，我們會再次看到它們的重要性，因為我們經歷的人口成長衝擊規模實在太大了。And then during the pandemic, a lot of other things started to matter, mostly related to the pandemic, and then the massive economic support programs that were launched after that. But those—The connection between population aging, demographic growth, and interest rates didn't go away, right? It just, you know, other things occupy better attention. But those pathways are still valid, and I think that we'll see in coming years that they matter again because of the size of the shocks that we've experienced to population growth.

13:25另一件我認為會影響利率的因素是財政赤字的改善。所以，你知道，如果你比較去年第二季到第四季的財政赤字，和前一年第二季到第四季的財政赤字，當然我認為主要的差異，最大的差異是關稅。並不是全部都是關稅，但最大的差異來自關稅。財政赤字大幅縮減，而我認為以年化來看，大概是——450，但那是4500億美元，不過這是我憑印象說的，所以可能不太正確。Another thing that I think weighs on the interest rate is the improving fiscal deficit. And so, you know, I think that if you look at the fiscal deficit in Q2 through Q4 of last year versus the fiscal deficit of Q2 through Q4 of the previous year, and of course I think the major delta is, the biggest delta is tariffs. It's not all tariffs, but the biggest delta is tariffs. The fiscal deficit came in quite a bit, and I think at an annualized rate it's probably, if I remember correctly, about—450, but that for 450 billion, but that's off the top of my head, so that might not be correct.

14:02但如果你比較2025年第二季到第四季，也就是日曆年度2025年第二季到第四季，和2024年日曆年度第二季到第四季，這確實是一個顯著的下降。我認為主要的差異在於關稅，對吧？所以，財政借款的下降也是會影響中性利率的因素，這也是我認為會對中性利率產生影響的部分。所以，你知道，我對中性利率的看法是在我的同事們看法範圍的下緣，如果你參考SEP，但我並沒有超出這個範圍。But it's a significant decline if you look at Q2 through Q4 of 2025, calendar Q2 through Q4 of 2025 versus calendar Q2 through Q4 of 2024. And I think the major delta is tariffs, right? So that declining, the decline in fiscal borrowing is something that pushes around the neutral rate too, and that is something that I think we'll also weigh on it. So, you know, so my view of the neutral rate is towards the bottom of the range of where my colleagues' view is, if you look at the SEP, but I'm not outside of the range.

14:33監管鬆綁、供給面政策與經濟潛力

14:33所以你剛才提到的另一個主要且持久的趨勢，就是所謂的去管制化這個概念。就在上週，聯邦準備理事會發布了一份初步的諮詢請求，針對聯準會對於去管制化貢獻的觀點進行更新。我很好奇你能不能分享一下你的看法，以及你是怎麼思考這件事的。如果你去上經濟學課，你會學到供給和需求，對吧？如果你把供給維持不變，然後一腳踩下需求的油門，把需求推高，你就會得到通膨。如果你把供給推高，那麼你就不會有通膨，你可以用更少的資源生產更多的東西。So the other major durable trend you mentioned was this idea of deregulation. And just last week, the Federal Reserve put out some initial ask for consultation on this update to the Federal Reserve's perspective on how it contributes to that, you know, update on deregulation. I'm curious if you just provide your perspective on that and how you're thinking about it. So if you look, if you take an economics class, there's supply and demand, right? And if you hold supply constant and you slam your foot on the gas for demand and you push demand out, you get inflation. And if you push supply out, then you don't get inflation. You can produce more with less.

15:19我認為，你知道，網路上有個大家常用的說法叫做「跑得很熱」（running it hot），對吧？我覺得「跑得很熱」這個說法，其實是取決於你用的是什麼引擎。如果你開著一台馬力很小的車，想要時速六十英里，那你大概就是在「跑得很熱」。但如果你開的是一台馬力很大的車，有幾百匹馬力，你以時速六十英里行駛，其實就不算「跑得很熱」，對吧？那台車可以輕鬆應付這個速度。這就是把供給面往外推和限制供給面的差別。And I think, you know, one expression that people online use a lot is running it hot, right? I think that running it hot is, if you use that expression, you're conditioning upon what the engine is. And if you're trying to go 60 miles an hour in a car with a handful of horsepower, you know, you're probably running it hot. But if you're trying to go 60 miles an hour in a car with a large number of horsepower, hundreds of horsepower, you're not running it hot, right? That car can handle that speed. And this is the difference between pushing the supply side out versus restricting the supply side.

15:52如果你推動經濟的供給面並且供給快速成長，那麼需求也可以快速成長，而且這並不會造成通膨。如果你限制供給面，經濟無法產出足夠的供給來滿足需求，那你就會產生通膨。我認為這就是差別所在。而這也是像 AI 和放寬管制這類事情發揮作用的地方，因為它們讓經濟能夠用更少的資源生產更多東西。資本深化也是如此。如果你看看去年稅收立法中針對投資的激勵措施，像是設備的全額費用化以及研發的全額費用化。If you push the supply side of the economy out and supply is growing quickly, then demand can grow quickly too, and it's not inflationary. If you're holding the supply side in and the economy can't produce supply to meet demand, then you're going to get inflation. And I think that's the difference. And that's where things like AI and things like deregulation come into play, is because they allow the economy to produce more with less. And that's also true of capital deepening. If you look at the incentives for investment that were in part of the tax legislation last year, you know, the full expensing on equipment and full expensing in R&D.

16:27這些措施會激勵對生產性資本的投資，而當你擁有更多生產性資本時，你就能生產更多，對吧？所以我認為這一切都是一個連動的結果——當你說像是「讓經濟過熱」這樣的說法時，其實是非常不精確的，因為這種說法假設供給是固定的，認為你無法提升汽車的馬力。但如果你推動去管制化、推動人工智慧、推動資本深化，你其實是在提升汽車的馬力，你可以用更少的資源生產更多，因此需求可以成長得更快，而當你加速時，這個引擎也不會過熱。所以我認為這個比喻需要做一些調整，但是——These incentivize investment in productive capital, and when you have more productive capital, you can produce more, right? And so I think it's all a function of—so when you sort of say something like running it hot, it's wildly imprecise because it's a statement that sort of assumes that supply is constant, that you can't increase the horsepower of the car. But if you go through deregulation, if you go through AI, if you go through capital deepening, you're increasing the horsepower of the car and you can produce more with less, and therefore demand can grow faster and the engine doesn't heat up when you go fast. So I think that that metaphor needs some adjustment, but—

17:04當我思考這些供給面政策以及它們與貨幣政策的互動時，很多影響都是透過我們所謂的——有兩個主要的傳導管道。第一個是價格會發生什麼變化，對吧？我剛才有稍微提到過放寬管制這件事。如果你能用更少的資源生產出更多產品，你就能降低生產成本，而這會反映在消費者價格上。如果你在消除進入障礙，因為法規有時會成為進入障礙，而我在一月的法規演講中也有詳細討論過這點，你就會創造出更多競爭。更多的競爭會降低價格，會降低加價率，也會降低壟斷利潤。When I think about these supply-side policies and their interaction with monetary policy, a lot of it flows through what we call—so there's two primary channels. One is what happens to prices, right? I talked a little bit about that with deregulation. If you can produce more with less, you lower the cost of production that feeds through into consumer prices. If you're removing barriers to entry, because regulations can serve as barriers to entry, and I talked a lot about this in my January regulation speech, you're creating more competition. More competition will lower prices. It will lower markups. It will lower monopoly profits.

17:42我認為，供給面影響我們在貨幣政策上所關心事項的另一個管道，就是我們所謂的產出缺口。產出缺口是指潛在成長，也就是經濟在失業率處於自然水準時所能產出的最大值，與實際產出之間的差距。所以，如果失業率是8%，那經濟中就有大量未被利用的閒置資源。經濟產出低於潛力，還有很多可以生產的空間。The other channel through which I think a lot of the supply side of stuff affects things that we care about for monetary policy is through what we call the output gap. And the output gap is the difference between potential growth, what the economy can produce with, you know, unemployment at its natural level, and where actual output is. And so if you've got the unemployment rate at 8%, there's tons of unused slack in the economy. There's tons of unused resources. The economy is producing below potential. There's a lot more it can produce.

18:01如果你的失業率是3%，那代表你已經超過潛在產能，因為你正試圖以一種不會引發通膨的方式生產超過經濟本身所能承受的產出，因此你會因此產生價格壓力。現在，當你在思考供給面衝擊時，不論是AI、石油、法規，無論是哪一種衝擊，真正重要的是，你要思考這些衝擊到底是讓實際供給或實際GDP、還是潛在GDP往外推還是往內拉，以及它們的變動幅度誰比較大。If you've got the unemployment rate at 3%, you're above potential because you're trying to produce more than the economy can produce in a non-inflationary way, and you get price pressures as a result. Now, what matters when you think about shocks to the supply side, whether they're AI, oil, regulation, no matter what the shock is, what matters when you think about this is, is it pushing out or pulling in actual supply, or Sorry, actual GDP or potential GDP by more or less.

18:43所以，如果你遇到這樣的情況，而且很多人會有一個假設，就是實際產出和潛在產出在面對這些生產力衝擊時，會大致以相同的幅度變動。有時候我認為這是一個不錯的假設。舉例來說，如果你回想90年代Greenspan的生產力故事。當時，前主席Greenspan主張生產力的提升會帶來高GDP成長，而且不會引發通膨。這個論點很大一部分是在於潛在GDP也會隨之增加，而實際GDP也會跟著上升。And so if you have a situation like, and one assumption that a lot of people make is that actual and potential move out by roughly the same amount in response to a lot of these productivity shocks. There are times when I think that that's a good assumption. So, for example, if you think about the Greenspan productivity story in the 90s. Where, you know, former Chairman Greenspan was making the case that improvements in productivity were going to lead to high GDP growth that was non-inflationary. A lot of that bet was that potential GDP was going to increase too, and that actual GDP would increase with it.

19:20因此，產出缺口並沒有變化，因為實際產出和潛在產出大致上是以相同的幅度上升。因此，政策其實不需要做出太大的反應。我認為這個論點在電信革命的背景下非常有道理，因為如果你要建設大量的電信容量，你必須挖開馬路、鋪設電信線路，然後再把馬路復原。這會帶來大量的投資活動，因此你會因為投資活動而看到需求的增加。同時，由於網路連接提升帶來的生產力提升，你也會看到供給的增加。And so therefore there's no change in the output gap because actual and potential are moving up by roughly the same amount. And therefore policy doesn't really need to respond. Now, I think that argument, I think, makes a lot of sense in the context of the telecoms revolution, because if you're building lots of telecom capacity, you have to tear up streets, lay telecom wire, put the streets back together. There's a lot of investment activity that occurs, and so you get the increase in demand because of the investment activity. And you get the increase in supply because you have a productivity enhancement from the increased internet connectivity.

19:58產出缺口的淨變化，我認為假設它可能為零是合理的。抱歉，應該說它可能是中性的。對於其他類型的衝擊，情況可能就不是這樣了。當然，針對法規衝擊的情況，我認為實際GDP的變動會遠小於潛在GDP的變動。所以，舉個例子，如果你有一座煙囪，在嚴格的碳排放法規下，你一天只能運作八小時，然後當碳排放法規放寬後，你可以一天運作十六小時，你不需要額外投資，因為你的資本存量已經在那裡了。你已經有那座煙囪了。The net change in the output gap, I think it's a fair assumption to assume it could be zero. Sorry, it could be neutral. With other types of shocks, that might not be the case. Certainly, with the case of a regulation shock, I think that actual GDP will move by a lot less than potential GDP. So, for example, if you've got a smokestack that you can run eight hours a day with strict carbon regulations, and then the carbon regulations ease and you can run it 16 hours a day, you don't need to do additional investment because you've got the capital stock already there. You've got the smokestack.

20:28只是你可以讓煙囪一天運作更多小時。因此，這是一個潛在GDP大幅上升的案例，我認為，實際GDP的增幅則沒有那麼大。你能夠生產的供給量大幅增加了，但需求的增幅並不大，因為你所做的只是放寬了生產限制，並沒有大幅提升投資需求。至於AI，我不確定。AI可能介於兩者之間。當然，AI確實創造了大量的投資需求，像是GPU和資料中心。我認為，許多GPU的需求其實是流向海外的。It's just that you can run the smokestack more hours a day. And so that's a case in which potential GDP goes up, I think, by a lot more than actual GDP. The amount of supply you can produce has gone up by a lot. But the amount of demand has not gone up by a lot because all you've done is ease production constraints. You haven't increased a lot of investment demand. AI, I don't know. AI could be somewhere in between. So certainly AI creates a huge amount of investment demand in the form of, you know, GPUs and, you know, data centers. A lot of the GPU demand, I think, is leaked overseas.

21:05你知道，其實很多GPU並不是——它們不是在美國製造的，對吧？當然，資料中心的需求確實會帶動美國的投資活動。當然，我們也確實支付了大量員工薪酬給那些從事AI開發、將AI整合到現有公司工作流程中的人員。但這對潛在GDP與實際GDP的影響會如何呢？我其實沒有一個明確的看法。You know, a lot of the GPUs are not—they're not made in the United States, right? Certainly data center demand does lead to investment activity in the United States. Certainly we pay a lot of employee compensation to people who are working on developing AI, integrating AI into existing company workstreams. But where does it shake out on potential GDP versus actual GDP? I don't really have a firm view.

21:29數位資產、瘦身主帳戶與金融創新

21:29所以我想在這裡換個主題，進入這場會議真正的核心，也就是數位資產以及你對它們的看法。我想先從這個「精簡主帳戶」的介紹開始談起，這個制度已經開始推行，我們看到 Kraken 已經獲准取得一個。我很想聽聽你對這些精簡主帳戶的看法，特別是你怎麼看它們與完整主帳戶之間的定位和角色。 當然可以。這個問題其實和我們剛剛的討論很有關聯，因為我們剛剛談到生產力成長，也談到科技進步。So I want to shift gears here into the heart of what we're truly here for this conference, which is digital assets and your perspective on them. And I want to start with this introduction of these skinny master accounts that have been begun. Rolling out, we've seen Kraken get approved for one. And so I would just love to get your perspective on how you think about the role of these skinny master accounts with respect to where it sits in comparison to, you know, a full-fledged master account. Sure. So this ties into the previous conversation nicely because we were talking about productivity growth and we were talking about technological advancements.

22:04如果你回顧人類歷史上非常非常長遠的時間來看，推動我們所經歷過所有人類繁榮成長的，其實就是科技進步。我認為這一點適用於各種各樣的科技進步。我也認為金融創新是其中一個重要的部分，因為金融創新，以及金融科技和金融能力的進步，有助於把資本分配到需要的地方，讓經濟能夠有效率地運作、生產更多，並且——總體來說推動人類的繁榮。所以我確實認為金融創新也很重要。And if you look over the very, very long term of human history, it is technological advances that drive all of the growth in human prosperity that we've experienced. And I think that's true of a wide variety of technological advances. And I think that financial innovation is an important part of that, because financial innovation and advances in financial technology and financial capabilities help to allocate capital to where it needs to be to make the economy produce efficiently, produce more, and—Just generally drive human prosperity. So I do think that financial innovation is important too.

22:37我認為為穩定幣引進這種精簡主帳戶（skinny master accounts）將會是推動這類創新發生的重要一步。聯邦準備理事會最近發布了一份徵求意見書，針對精簡主帳戶徵求意見。我們收到了大量的意見回饋。現在工作人員正在審查這些意見。我自己也看過其中幾份意見。有些讓我印象深刻的點，包括像是ACH存取權限，以及對帳戶餘額規模、精簡主帳戶規模的上限設置。但這是一個正在積極檢討和制定規則的領域。所以，這件事確實正在推進。我很期待由負責支付委員會的Waller理事來推動這件事進入下一階段，我認為這非常有潛力。I think that introducing these skinny master accounts for stablecoins will be an important step in the direction of allowing that type of innovation to occur. The Federal Reserve recently put out a request for information for comments on skinny master accounts. And we received a large volume of these comments. The staff are going through the comments now. I've looked at a few of the comments. Some of the things that stuck out at me were things like ACH access and caps on the size of the balances, the size of the skinny master accounts. But this is an area of active examination and active rulemaking. And so, you know, it is something that is moving ahead. I'm excited for Governor Waller, who leads the Payments Committee, to sort of push this to next steps, and I think it's got a lot of potential.

23:35穩定幣、全球資金流與美元體系

23:35我們目前看到的另一個重大趨勢，當然就是穩定幣。你最近針對穩定幣發表了一場具有里程碑意義的演說，我很想聽聽你現在對這個議題的想法是如何演變的，以及你認為穩定幣在更廣泛金融體系中的角色，你認為它會被納入哪些領域，你認為它如何補足現有的體系，還有，就是你現在對這個議題的整體思考。謝謝。是的，我在十一月發表了一場演說。大概是那個時候，沒錯。主題就是穩定幣，你知道，我在那場演說中針對穩定幣提出了幾個論點。我主要的論點是，如果你擁有符合GENIUS Act的穩定幣，它們能幫助你，能促進數位支付的發展。The other big trend that we're seeing is, of course, around stablecoins. You had a landmark speech about stablecoins recently, and I would love to just hear about how you're evolving your thinking there and its role within the broader financial system, where you see it being incorporated, where you see it complements what already exists there, and yeah, just how you're thinking about that. Thanks. So I gave a speech in November. Something like that, yeah. About stablecoins, and, you know, I made a couple of arguments about stablecoins in that speech. The major argument that I made was that if you've got Genius Act-compliant stablecoins, they help you, they help facilitate digital payments.

24:24但從資本儲蓄池的角度來看，在美國、在歐洲、在那些已經能夠取得以美元計價的儲蓄帳戶、以美元計價的貨幣市場基金、國庫券的地方，如果你有一個資本開放、資本市場開放的體系，從儲蓄的角度來說，我認為穩定幣並沒有帶來太多額外的價值。從支付的角度來看，穩定幣確實有其貢獻，因為它們促進了數位支付。但如果你已經擁有一個龐大的資本池，我不認為穩定幣能帶來太多額外的效益。當然，在有資本管制的地方，或是在因地理或發展因素導致銀行體系難以滲透的地區，情況就不是這樣了。But from a savings pool of capital perspective, in the United States, in Europe, in places that already have access to dollar-denominated savings accounts, dollar-denominated money market funds, Treasury bills, if you have a system with open capital, open capital markets. From a savings perspective, stablecoins don't, I think, add as much. From a payments perspective, they do, because they're facilitating digital payments. But if you've got a huge pool of capital, I don't know that stablecoins are adding that much. Now, that's not true in places with capital controls, and it's not true in places where the banking system struggles to penetrate because of geography or development, right?

25:10因此，世界上有很大一部分地區的人們生活在資本管制之下，如果他們想要取得美元儲蓄工具，例如美元存款、國庫券、貨幣市場基金，他們根本沒有能力這麼做，因為他們所在國家的法律規定，你就是不能把你的錢轉換成美元。我們就是不會讓你這麼做。無論這些人多麼渴望取得美元，他們就是被禁止。還有其他地區非常偏遠，可能根本沒有銀行服務。And so you've got large parts of the world where people are living behind capital controls, where if they wanted access to dollar savings instruments like dollar deposits, treasury bills, money market funds, they don't have the capacity to do so because the country they're living in, the law is you just can't move your money into dollars. We're not going to let you. And it doesn't matter how bad the people want access to dollars, they're just not allowed to. There are other parts of the world where it's very rural and there may not be banking services.

25:39也許人們有一支手機，所以他們可以透過手機連上衛星網路，但他們其實無法使用真正的銀行服務。或者該國的銀行服務過於不可靠且波動太大，或者通貨膨脹太嚴重，讓他們不想使用銀行服務。所以我的看法是，我對於穩定幣的成長非常樂觀，但我認為，很多穩定幣的採用其實會來自於那些希望以美元計價、希望資金停留在美元、但目前無法停留在美元的大量資金。And maybe people have a cell phone, and so they, you know, have a satellite link on their cell phone, but they have no access to actual banking services. Or the banking services in the country are too unreliable and volatile, or there may be too much inflation, and they don't want to use the banking services. So my perspective is that I'm very optimistic about stablecoin growth, but I think that a lot of the stablecoin uptake is actually going to come from large volumes of money that want to be denominated in dollars, that want to sit in dollars, and are currently unable to sit in dollars.

26:13沒錯，有大量的儲蓄池，原本沒有其他方式能進入美元體系。突然之間，現在有了一種方法可以做到這件事。當然，你仍然需要進入加密貨幣生態系統才能這麼做。所以這裡仍然存在一個進入管道的問題，但我認為這讓問題變得比較容易解決。我有點把這件事想像成，像是叫車軟體這種新科技，打破了計程車牌照的壟斷。這種事情創造了讓人們開始繞過過去無法取得美元存款的障礙的可能性。Right, huge pools of savings that have no other way of getting into the dollar system. Suddenly there's a way of doing it. And of course, you still need to get onto the crypto ecosystem to do that. So there is sort of still an entry rail issue, but I think it makes the problem easier. And I sort of think of it a bit like, you know, a rideshare app sort of being a new technology that broke up the monopoly of a taxi cab medallion, right? It's that type of thing that sort of creates that ability to sort of start circumventing barriers to dollar deposits that people didn't have access to before.

26:46當然，將這個話題拉回到我們剛才討論的中性利率，如果來自世界其他地區的大量資金流入以美元計價的儲蓄，這將會對中性利率產生壓力。對吧。這正是發生在1990年代末期和2000年代初期的情況，前任主席Ben Bernanke稱之為全球儲蓄過剩。如果我們達到對穩定幣成長較為樂觀的預測，你可能會看到的規模，雖然可能沒有全球儲蓄過剩那麼大，但假設有一半那麼大。Now, of course, tying this back to the neutral rate that we were discussing before, if you have huge inflows from the rest of the world into U.S. dollar-denominated savings, that's going to weigh in the neutral rate. Right. This is what happened in the late '90s and the early 2000s, what former Chairman Ben Bernanke called the global savings glut. If we hit the more optimistic projections of stablecoin growth, you could be looking at magnitudes that are, you know, maybe not quite as big as the global savings glut, but let's say half as big.

27:18所以，你知道，這些事情確實有可能對貨幣政策產生非常重大的影響，而這當然會是一個非常強大的力量，對利率產生壓力，使其維持在較低的水準。So, you know, there is the scope for these to matter, you know, very substantially for monetary policy as well, which of course would be a very, you know, which would be a very powerful force weighing on interest rates, keeping them lower.

27:38代幣化存款與金融服務創新

27:38我很好奇，你怎麼看待這個等式的另一面，我覺得這一點常常被低估，就是代幣化存款？你對這個有什麼看法？ 嗯，對我來說，代幣化存款讓我覺得——而且我要說清楚，我並沒有特別深入研究過代幣化存款。不過就我所知，可能還比你了解得少，這讓我覺得它是在現有銀行已經提供的服務上做出的一種改進，對吧？這最終會不會徹底革新銀行體系呢？I'm curious, how do you think about the other side of the equation, which I think goes understated, which is tokenized deposits? What do you think about those? Yeah, so tokenized deposits strike me as a—and to be clear, I have not made an extensive study of tokenized deposits in particular. But what I know about them, which is probably less than you do, is that they strike me as an improvement on services that are already being offered by banks, right? Is that something that ultimately ends up revolutionizing the banking system?

28:14我不知道，不過這種代幣化存款給我的感覺是，呃，這就像是銀行運用科技來提升其金融服務的一連串進步中的又一步。不過我並沒有深入研究這個議題，如果你告訴我一些新的資訊，我也願意改變我的看法。I don't know, but this tokenized deposit strike me as, as, as a, you know, another step in a long line of improving the financial services that banks offer by utilizing technology. But I haven't made an extensive study of it, and I would be open to changing my mind based on things you tell me.

28:38給美國金融創新者的建議與鼓勵

28:38對於那些在加密產業中扮演核心角色的人們，以及你知道的，潛在的創辦人等等，正在努力推動這股龐大的金融創新趨勢，我們很想聽聽你們對於那些試圖在美國金融體系中建構和創新的人的臨別贈言。那我就重申我之前說過的話：創新是推動人類長期繁榮的主要動力。所以，如果你正在從事創新，感謝你所做的一切。Just for the folks out there who are core to the crypto industry and, you know, potential founders, et cetera, that are working towards this huge, you know, tailwind that is going on of financial innovation, we'd just love to hear any sort of parting words to those that are trying to build and innovate in the space in the U.S. financial system of, yeah, just your final words to them. So I'll repeat what I said before, which is that innovation is the driver, is the main driver of long-run human prosperity. So if you are working on innovation, thank you for doing what you do.

29:11關於支付及支付技術相關的議題，你知道，當聯邦準備理事會制定政策時，我們會遵循《行政程序法》，這是所有監管機構都必須遵守的規定。我們會發布所謂的擬議規則制定通知，也會發布徵求資訊的請求。請回應這些通知，給我們意見。我們會尋求來自產業、利害關係人以及創新者的意見，這樣我們才能知道現行法規是否發揮了應有的作用，或者哪些地方需要推動或調整，才能讓法規更適合促進經濟效率。因此，我們會發布這些通知，讓大家告訴我們，哪些地方需要改變，才能讓經濟運作得更有效率。And with respect to issues in payments and payments technologies, you know, when the Federal Reserve makes policy, we follow. The Administrative Procedure Act, which is what all regulatory agencies have to do, and we issue, we issue, you know, sort of notices of proposed rulemaking. We issue requests for information. Respond to those. Give us comments. You know, we search for comments from industry, from stakeholders, from innovators, so that we can know whether the regulations are doing the job that they are supposed to or where they need to be pushed or pulled. To get them into a place to make the economy more efficient. So, you know, we put out these notices, you know, sort of let us know what needs to change to make the economy more efficient.

29:53太棒了，非常感謝你來到數位資產高峰會分享，Miran 理事。真的很精彩，謝謝你。謝謝邀請我來。Incredible. Well, can't thank you enough for coming here and talking at the Digital Asset Summit, Governor Miran. That was incredible. Thank you. Thanks for having me.