Posted by AGORACOM
at 1:57 PM on Tuesday, January 28th, 2020
SPONSOR: Lomiko Metals is focused on the exploration and development of minerals for the new green economy such as lithium and graphite. Lomiko owns 80% of the high-grade La Loutre graphite Property , Lac Des Iles Graphite Property and the 100% owned Quatre Milles Graphite Property. Lomiko is uniquely poised to supply the growing EV battery market. Click Here For More Information
According
to research by BloombergNEF, European automakers and governments will
move toward helping curb global warming with stricter carbon emissions
regulations, which could force an electric-vehicle revolution.
In the United States, electric vehicles are primarily being purchased
by consumers that want to take action on their own. Fuel is cheap, the
country doesn’t have a real climate change plan, and large vehicles like
pickups are king. All of this means that there’s little incentive,
beyond the $7,500 federal tax credit, to purchase an EV. That, though,
isn’t the case in other countries like China and, soon to be Europe.
EV Revolution Coming This Year
According
to a report by Bloomberg and a forecast from BloombergNEF, Europe will
see an electric revolution in 2020. The outlet states that the country’s
government will soon look to cut carbon emissions from vehicles as part
of a plan to curb global warming. This, in turn, will force automakers
to introduce electric vehicles.
Bloomberg claims that sales of
electric cars are set to increase to 2.5 million units in 2020. That
figure represents an increase of 20 percent from 2019.
Just like this year, China will continue to lead the way forward for sales. But the country recently decided to reduce subsidies for EV owners,
which could help Europe gain a larger piece of the market. The outlet’s
forecasting claims that Volkswagen’s push to become an electric-vehicle
force will boost the number of electrified vehicles in Europe. In
total, the outlet expects 800,000 electric cars to be sold in Europe in
2020.
“The long-term future is really bright, but in the short
term we’re expecting growth to be relatively slow,” said Colin
McKerracher, an analyst at BloombergNEF. “You’re still in the middle of
this transition, from a market driven by direct subsidies toward one
driven by a combination of real consumer demand and other big policy
mechanisms.”
Better Prices, More Infrastructure Coming
Another
important aspect of electric vehicles that will help sales increase in
Europe are decreasing lithium-ion battery prices. The outlet states that
prices per kilowatt-hour will hit roughly $135 – approximately 13
percent lower than in 2019. With the increase of battery production,
better battery designs, and more sales, battery prices are expected to
tumble.
All of these things mean that more chargers will be
needed. Luckily, public chargers are expected to rise to 1.2 million, up
from 880,000 last year. The increase in chargers will come in part from
governments and energy companies looking to expand infrastructure to
support the increase in demand for electric cars.
Another
interesting trend to look at in 2020 include other forms of electrified
transportation. A few companies, even automakers, showcased flying electric cars at CES.
While it’s unlikely that one would come out in 2020, it’s likely
something that more companies will pursue this year. Other forms of
transportation, including boats could go electric in 2020, too.
Posted by AGORACOM
at 12:26 PM on Tuesday, January 21st, 2020
SPONSOR: Lomiko Metals is focused on the exploration and development of minerals for the new green economy such as lithium and graphite. Lomiko owns 80% of the high-grade La Loutre graphite Property , Lac Des Iles Graphite Property and the 100% owned Quatre Milles Graphite Property. Lomiko is uniquely poised to supply the growing EV battery market. Click Here For More Information
While many people consider Detroit home of the automobile, the
southeast region of the U.S. is becoming a hotbed for auto
manufacturing. Automakers BMW,
Mercedes Benz, Volvo, Toyota, Honda and Hyundai built assembly plants
in the region to manufacture vehicles for the U.S. and global markets.
Most recently, Toyota and Mazda recently announced they will be
opening a new $1.6 billion plant in Huntsville, Alabama, adding around
4,000 new jobs to the region. Now Volvo becomes the latest automaker to
expand its U.S. manufacturing with a new electric vehicle battery plant.
The automaker announced plans to build an electric battery plant at
its assembly plant in Ridgeville, South Carolina to support the launch
of electrified Volvo models for the U.S. market. Construction of the
battery assembly plant will be completed by the end of 2021, a Volvo
spokeswoman said to Automotive News.
The battery production plant is part of a previously announced $600
million project that is already underway at Volvo’s plant in Ridgeville,
S.C., which includes adding a second production line and Volvo Car
University. The 2.3 million sq. ft. facility includes a body shop, paint
shop, final assembly, a vehicle processing center and an office
building.
The Ridgeville plant is Volvo’s first in the U.S. Construction began in 2015.
At that facility, employees will assemble and test the lithium ion
battery packs that will power the electric XC90. By assembling the packs
on at the plant, Volvo hopes to reduce shipping costs involved in
transporting the heavy batteries.
Dallas Bolen, a manager with Volvo’s product launch group, told local
media outlet the Post and Courier that local battery production would
be more cost-effective than building batteries off-site then having to
transport them to the factory.
The Ridgeville plant is currently the production home of the Volvo
S60 sedan. The U.S.-built S60s are exported around the world through the
Port of Charleston, one of the busiest ports in the U.S.
Volvo’s next EV will be the XC40 Recharge. It will arrive at U.S. dealers later this year.
The South Carolina plant will become the global production center for
the third-generation XC90 flagship crossover. Volvo plans to build the
next generation XC90 sport utility vehicle in 2022, along with a
fully-electric version. The plant has the capacity to build 150,000
vehicles annually.
Volvo has not said how much of the XC90’s production at the $1.1
billion factory will be devoted to the battery-electric variant.
That next-generation XC90 will be built on the next version of
Volvo’s Scalable Product Architecture platform, referred to as SPA2. The
new electric vehicle architecture is designed to make it easy to add
new technology, such as microprocessors, sensors and camera technology.
Volvo declined to release its production capacity for the battery
assembly plant or say how many jobs it will create. Overall, the planned
XC90 production line is expected to create about 1,000 jobs.
The XC90 would be Volvo’s third battery-powered model following the
electric version of the popular XC40 compact crossover, was unveiled in
October.
The electric XC40
is expected to arrive in U.S. dealerships in the fourth quarter of
2020. The crossover will be competitively priced under $48,000, after
the $7,500 federal tax credit, Volvo said.
The new battery plant will support Volvo’s push to electrify around
half of its lineup. The automaker aims for EVs to account for half of
its global sales by 2025. Over the next five years, Volvo expects to
launch a fully electric vehicle every year.
“A Volvo built in 2025 will leave a carbon footprint that is 40
percent lower than a car that we build today,” Volvo CEO Hakan
Samuelsson said during a press event in October. “We made safety part of
the brand. We should do the same with sustainability.”
In November 2019, Volvo Cars announced it will be the first carmaker
to implement global traceability of cobalt used in its batteries by
applying blockchain technology,
ensuring that customers can drive battery-powered Volvos knowing the
raw materials for the batteries has been responsibly sourced.
Posted by AGORACOM
at 12:49 PM on Sunday, January 19th, 2020
Sponsor: Affinity Metals (TSX-V: AFF) a Canadian mineral exploration
company building a strong portfolio of mineral projects in North
America. The Corporation’s flagship property is the Drill ready Regal
Property near Revelstoke, BC. Recent sampling encountered bonanza grade
silver, zinc, and lead with many samples reaching assay over-limits.
Further assaying of over-limits has been initiated, results will be
reported once received. Click Here for More Info
From the HRA Journal: Issue 314
The fun doesn’t stop. Waves of liquidity continue to wash traders
cares away. Even assassinations and war mongering generate little more
than half day dips on Wall St. It seems nothing can get in the way of
the bull rally that’s carrying all risk assets higher.
It feels like it could go on for a while, though I think the
liquidity will have to keep coming to sustain it. By most readings,
bullishness on Wall St is at levels that are rarely sustained for more
than a few weeks. Some sort of correction on Wall St seems highly
likely, and soon. Whether its substantial or just another blip on the
way higher remains to be seen.
The resource sector, especially gold and silver stocks, have had
their own rally. Our Santa Claus market was as good or better than Wall
St’s for a change. And I don’t think its over yet. I think we’re in for
the best Q1 we’ve seen for a few years. And we could be in for something
better than that even. I increasingly see signs of a major rally
developing in the gold space. It’s already been pretty good but I think a
multi-quarter, or longer, move may be starting to take shape.
I usually spend time on all the metals in the first issue of the
year. But, because the makings of this gold rally are complex and long
in coming I decided to detail my reasoning. That ended up taking several
pages so I’ll save talk on base metals and other markets for the next
issue.
No, I’m not writing about Louis IV, though there might be some
appropriateness to the analogy, now that I think about it. The quote is
famous, even though there’s no agreement on what it was supposed to
mean. Most figure Louis was referring to the biblical flood, that all
would be chaos once his reign ended.
The deluge I’m referring to isn’t water. It’s the flood of money the
US Fed, and other central banks, continue to unleash to keep markets
stable. Markets, especially stock markets, love liquidity. You can see
the impact of the latest deluge, particularly the US Fed’s in the chart
below that traces both the SPX index value and the level of a “Global
Liquidity Proxy†(“GLPâ€) measuring fiscal/monetary tightness and
weakness.
You can see the GLP moved lower in late 2018 as the Fed tightened and
the impact that had on Wall St. Conversely, you can see the SPX running
higher in the past couple of months as the US backed off rate
increases, increased fiscal deficit expansion, and grew the Fed balance
sheet through, mainly, repo market operations.
Wall St, and most other bourses, are loving these money flows. The
Santa Claus rally discussed in the last issue continued to strengthen
all the way to and through year end. As it turned out, the Fed either
provided enough backstop in advance or the yearend repo issues were
overstated. The repo market itself was calm going through year end and a
lot of the short-term money offered by the Fed during that week wasn’t
taken down.
Everything may have changed in the past couple of days with the
dramatic increase in US-Iran tensions. I don’t know how big an issue
that will be, since no one knows what form Iran’s retaliation will be or
how much things will escalate. I DO think it’s potentially a big deal
with very negative connotations, but it may take time to unfold. Someone
at the Fed thought so too, as the past couple of days saw a return to
large scale Fed lending in the repo market.
I’ve no doubt Iran will try and take revenge for the assassination of
its most famous military commander by the US. But I don’t know what
form it will take and if this means the US has drawn itself into the
Mideast quagmire even more. I fear it has though. The US is already
talking about adding 3,000 troops to its Mideast presence and they’re
just warming up. Even larger scale attacks, if they happen, may not
derail Wall St, but they’re certainly not a positive development at any
level.
We know how stretched both market valuations and sentiment were
before the Suleimani drone strike. The chart below shows a three-year
trace of the “fear/greed indexâ€. You can see that its hardly a stable
reading. It flip flops often and extreme readings rarely hold for long.
At last check, the reading was 94% bullish.
Sentiment almost never gets that bullish and, when it does, nothing
good comes of it for bulls. A reading that close to 100% tells you we’re
just about out of buyers. Whatever happens in and around Iran, I think a
near term correction is inevitable. The only question is whether it’s a
large one or not.
A rapid escalation in US-Iran tensions could certainly make a near
term correction larger. If the flood of liquidity continues though, a
correction could just be another waystation on the road to higher highs.
There are a couple of other dangers Wall St still faces that I’ll touch
on briefly at the end of this article. First however, lets move on to
the main event for us-the gold market.
It wasn’t just the SPX enjoying a Santa rally this year. Gold
experienced the rally we were hoping for that gold miner stocks seemed
to be foretelling early last month. Gold’s been doing well since it
bottomed at $1275 in June, but it didn’t feel that way during the long
hiatus between the early September high and the current move. The gold
price currently sits above September’s multi-year high, after breaching
that high in the wake of the Baghdad drone strike. And the first
retaliatory strike by Iran. Volatility will be very high for a while
going forward.
I think we’ll see more multi-year highs going forward. I hate that
the latest move higher is driven by geopolitics. Scary geopolitics and
military confrontations mean people are dying. We don’t want to profit
from misery. And we won’t anyway, if things get ugly enough in the
Mideast to scare traders out of the market.
Geopolitical price moves almost always unwind quickly. I’d much
prefer to see gold moving higher for macro reasons, not as a political
safety trade. I expect more political/military inspired moves. As the
Iran conflict unfolds. Make no mistake, Iran is NOT Iraq. Its army is
far larger, better trained and better equipped than Iraq. This could get
ugly.
The balance of this piece will deal with my macro argument for higher
gold prices over an extended period. The geopolitical stuff will be
layered on top of that for the next while and could strengthen both gold
prices and the $US in risk-off trading. It should be viewed as a
separate event from the argument laid out below.
What else is driving gold higher? In part, it was gold’s inverse
relationship with the US Dollar. As you already know, I’m not a believer
that “its all about the USD, all the time†when it comes to the gold
market. That’s an over-simplification of a more complex relationship. It
also discounts the idea of gold as its own asset class that trades for
its own reasons.
If you look at the gold chart above, and the USD chart below it, its
immediately apparent that there isn’t a constant negative correlation at
play. Gold rallied during the summer at the same time the USD did and
for the same reason; the world-wide explosion of negative real yields.
Gold weakened a bit when yields reversed to the upside and the USD got a
bit of traction, but things changed again at the start of December.
The USD turned lower and lost two percent during December. US bond
yields were generally rising during the month and the market (right or
wrong) was assuming economic growth was accelerating. So, neither of
those items explains the USD weakness.
If gold was a “risk off†trade, you sure couldn’t see it in the way
any other market was trading. So, is there another explanation for
recent strength in the gold price, and what does it tell us about 2020
and, perhaps, beyond?
Well, I’ve got a theory. If I’m right, it could mean a bull run for gold has a long way to go.
Some of this theory will be no surprise to you because it does
partially hinge on further USD weakness. There are long term structural
reasons why the US currency should weaken. But there are also
fluctuating sources of demand for USDs, particularly from offshore
buyers and borrowers that transact in US currency. That can create
enough demand to strengthen the US over long periods. We just went
though one such period, but it looks like that may have come to an end,
with more bearish forces to the USD reasserting themselves.
How did we get here? Let’s start with the big picture, displayed on
the top chart on the next page. It gives a long-term view of US Federal
deficits and the unemployment rate. Normally, these travel in tandem.
Higher unemployment means more social spending and higher deficits.
Government spending expands during recessions and contracts-or should-
(as a percentage of GDP) during expansions. Classic Keynesian stuff.
You rarely see these two measures diverge. The two times they did
significantly before, on the left side of the chart, was due to “wartime
deficits†which acted (along with conscription) to stimulate the
economy and drive down unemployment.
You can see the Korean and Vietnam war periods pointed out on the chart.
The current period stands out for the extreme size of the divergence.
US unemployment rates are at multi decade lows and yet the fiscal
deficit as a percentage of GDP keeps rising. There has never been a
divergence this large and its due to get larger.
We know why this is. Big tax cuts combined with a budget that is
mostly non-discretionary. And the US is 10 years into an economic
expansion, however weak. Just think what this graph will look like the
next time the US goes into recession.
We can assume US government deficits aren’t going to shrink any time
soon (and I think we can, pun intended, take that to the bank). That
leaves trade in goods to act as a counterbalance to the funding demand
created by fiscal deficits.
The chart above makes it clear the US won’t get much help from
international trade. The US trade balance has been getting increasingly
negative for decades. It’s better recently, but unlikely to turn
positive soon, and maybe not ever.
To be clear, this is not a bad thing in itself, notwithstanding the
view from the White House. The relative strength of the US economy and
the US Dollar and cheaper offshore production costs have driven the
trade balance. It’s grown because Americans found they got more value
buying abroad and the world was happy to help finance it. It’s not a bad
thing, but not a US Dollar support either.
The more complete picture of currency/investment flows is given by
changes in the Current Account. In simplified terms, the Current Account
measures the difference between what a country produces and what it
consumes. For example, if a country’s trade deficit increases, so does
its current account deficit. If there are funds flowing in from overseas
investments on the other hand, this decrease the Current Account
deficit or increase the surplus.
The graph below summarizes quarterly changes in the US current
account. You can see how the balance got increasingly negative in the
mid 2000’s as both imports and foreign investment by US companies
increased.
Not coincidentally, this same period leading up to the Financial
Crisis included a sustained downtrend in the US Dollar Index. The USD
index chart on the bottom of the next page shows the scale of that
decline, from an index value of 120 at the start of 2002 all the way
down to 73 in early 2008.
The current account deficit (and value of the USD) improved markedly
up to the end of the Financial Crisis as money poured into the US as a
safe haven and consumers cut back on imports. The current account
deficit bas been relatively stable since then, running at about
$100bn/quarter until it dipped a bit again last year.
Trade, funds flows and changes in money supply have the largest
long-term impacts on currency values. When the US Fed ended QE and
started tightening monetary conditions in 2014, the USD enjoyed a strong
rally. The USD Index was back to 100 by early 2015 and stayed there
until loosening monetary conditions-and lots of jawboning from
Washington-led to pullback. Things reversed again and the USD maintained
a mild uptrend from early 2018 until now.
There are still plenty of US Dollar bulls around, and their arguments
have short-term merit. Yes, the US has higher real interest rates and
somewhat higher growth. Both are important to relative currency
valuations as I’ve said in the past. Longer term however, the “twin
deficits†-fiscal and current account-should underpin the fundamental
value of the currency.
Movements don’t happen overnight, especially when you’re talking
about the worlds reserve currency that has the deepest and largest
market supporting it. Changing the overall trend for the USD is like
turning a supertanker. I think it’s happening though, and it has big
potential implications for commodities, especially gold.
Dollar bulls will tell you the USD is the “cleanest shirt in the
laundry hamperâ€, referring to the relative strength of the growth rate
and interest rates compared to other major currencies. That’s true if we
just look at those measures but definitely not true when we look at the
longer term-fiscal and current account deficits.
In fact, the US has about the worst combined fiscal/current account deficit in the G7. The chart at the bottom of this page, from lynalden.com
shows the 2018 values for Current Account and Trade balances for a
number of major economies, as a percentage of their GDP. It’s not a
handsome group.
Both the trade and current account deficits are negative for most of
them. In terms of G7 economies, the US has the worst combined
Current/Trade deficit at 6% of GDP annually. You may be surprised to
note that the Current/Trade balance for the Euro zone is much better
than the US, thanks to a large Trade surplus. Much of that is generated
by Germany. Indeed, this chart explains Germanys defense of the Euro.
It’s combined Trade/Current Account surplus is so large it’s currency
would be skyrocketing if it still used the Deutschmark.
Because the current account deficit is cumulative, the overall
international investment position of the US has continued to worsen. The
US has gone from being an international creditor to an international
debtor, and the scale if its debt keeps increasing. That means it’s
getting harder every year to reverse the current account position as the
US borrows ever more abroad to cover its trade and fiscal deficits.
Interest outflows keep growing and investment inflows shrinking.
Something has to give.
The US has to borrow overseas, as private domestic demand for
Treasury bonds isn’t high enough to fund the twin deficits. In the past,
whenever the US Dollar got too high, offshore demand for US government
debt diminished. It’s not clear why. Maybe the higher dollar made
raising enough foreign funds difficult, or perhaps buyers started
worrying about the USD dropping after they bought when it got too
expensive. Whatever the reason, foreign holdings of US Treasuries have
been declining, forcing the US to find new, domestic, buyers.
Last year, the US Fed stopped its quantitative tightening program,
due to concerns about Dollar liquidity. Then came the repo market. Since
September, the Fed’s balance sheet has expanded by over $400 billion,
mainly due to repo market transactions.
The Fed maintains this “isn’t QE†because these are very short duration transactions but, cumulatively, the total Fed balance sheet keeps expanding. The “QE/no QE†debate is just semantics.
What do these transactions look like? Mostly, its Primary Dealers,
banks that also take part in Treasury auctions, in the repo market. The
Fed buys bonds, usually Treasuries, from these banks and pays for them
in newly printed Dollars. That injects money into the system, helps hold
down interest rates in the repo market and, not coincidentally,
effectively helps fund the US fiscal deficit. To put the series of
transactions in their simplest form, the US is effectively monetizing its deficit with a lot of these transactions.
The chart below illustrates the problem for the Primary Dealer US
banks. They’ve got to buy Treasuries when they’re auctioned-that is
their commitment as Primary Dealers. They also need to hold minimum cash
balances as a percentage of assets under Basel II bank regulations.
Cash balances fell to the minimum mandated level by late 2019- the
horizontal black line on the chart. That’s when the trouble started.
These banks are so stuffed with Treasuries that they didn’t have
excess cash reserves to lend into the repo market. Hence the blow up
back in September and the need for the Fed to inject cash by buying
Treasuries. The point, however, is that this isn’t really a “repo market
issueâ€, that’s just where it reared its head. It’s a “too many
Treasuries and not enough buyers†problem.
It will be tough for the Treasury to attract more offshore buyers
unless the USD weakens, or interest rates rise enough to make them
irresistible. Or a big drop in the federal deficit reduces the supply of
Treasuries itself.
I doubt we’ll see interest rates move up significantly. I don’t think
the economy could handle it and it would be self-defeating anyway, as
the government deficit would explode because of interest expenses. And
that’s not even taking into account the fact that President Trump would
be freaking out daily.
Based on recent history and political expediency, I’d say the odds of
significant budget deficit reductions are slim and none. That’s
especially true going into an election year. There’s just no way we’re
going to see spending restraint or tax increases in the next couple of
years. Indeed, the supply of Treasuries will keep growing even if the US
economy grows too. If there is any sort of significant slowdown or
recession the Federal deficit will explode and so will the new supply of
Treasures. Not an easy fix.
Barring new haven demand for US Treasuries, odds are the Fed will
have to keep sopping up excess supply. That means expanding its balance
sheet and, in so doing, effectively increasing the US money supply.
That brings us (finally!) to the “money shot†chart that appears
above. It compares changes in the size of the Fed balance sheet and the
US Dollar Index. To make it readable and allow me to match the scales, I
generated a chart that tracks annual percentage changes.
The chart shows a strong inverse correlation between changes in the
size of the Fed balance sheet and the value of the USD. This is
unsurprising as most transactions that expand the Fed balance sheet also
expand the money supply.
It’s impossible to tell how long the repo market transactions will
continue but, after three months, they aren’t feeling very “temporaryâ€.
To me, it increasingly looks like these market operations are “debt
monetization in dragâ€.
I don’t know if that’s the Fed’s real intent or just a side effect.
It doesn’t really matter if the funding and money printing continues at
scale. Even if the repo market calms completely, the odds are good we
see some sort of “new QE†start up. Whatever official reason is given
for it; I think it will happen mainly to soak up the excess supply of
Treasuries fiscal deficits are creating.
I don’t blame the FOMC if they’re being disingenuous about it. That’s
their job after all. If you’re a central banker, the LAST thing you’re
going to say is “our government is having trouble finding buyers for its
debtâ€, especially if its true.
With no prospect of lower deficits and apparent continued reduction
in offshore Treasury holdings, this could develop into long-term
sustained trend. I don’t expect it to move in a straight line, markets
never do. A severe escalation in Mideast tensions or the start of a
serious recession could both generate safe-haven Treasury buying. Money
flows from that would take the pressure off the Fed and would be US
Dollar supportive too.
That said, it seems the US has reached the point where a substantial
increase in its central bank’s balance sheet is inevitable. Both Japan
and the Eurozone have gotten there before the Fed, but it looks like it
won’t be immune.
The Eurozone at least has a “Twin surplus†to help cushion things.
And Japan, considered a basket case economically, had an extremely deep
pool of domestic savings (far deeper than the US) to draw on. Until very
recently, Japan also ran massive Current Account surpluses thanks to
decades of heavy investments overseas by Japanese entities. Those
advantages allowed the ECB and especially the BoJ to massively expand
their balance sheets without generating a huge run up in interest rates
or currency collapse.
I don’t know how far the US Fed can expand its balance sheet before
bond yields start getting away from it. I think pretty far though.
Having the world’s reserve currency is a massive advantage. There is
huge built in demand for US Dollars and US denominated debt. That gives
the Fed some runway if it must keep buying US Treasuries.
Assuming a run on yields doesn’t spoil the party, continued balance
sheet and money supply expansion should put increasing downward pressure
on the US Dollar. I don’t know if we’ll see a move as large as the
mid-2000s but a move down to the low 80s for the USD Index over the
course of two or three years wouldn’t be surprising.
It won’t be a straight-line move. A recession could derail things,
though the bear market on Wall St that would generate would support
bullion. Currency markets tend to be self-correcting over extended
periods. If the USD Index falls enough and there is a bump in US real
interest rates offshore demand for Treasuries should increase again.
The bottom line is that this is, and will continue to be, a very
dynamic system. Even so, I think we’ve reached a major inflection point
for the US currency. The 2000s were pretty good for the gold market and
gold stocks. We started from a much lower base of $300/oz on the gold
price. Starting at a $1200-1300 base this time, I think a price above
$2000/oz is a real possibility over the next year or two.
It’s not hard to extrapolate prices higher than that, but I’m not
looking or hoping for those. I prefer to see a longer, steadier move
that brings traders along rather than freaking them out.
This prediction isn’t a sure thing. Predictions never are. But I
think the probabilities now favor an extended bull run in the gold
price. Assuming stock markets don’t blow up (though I still expect that
correction), gold stocks should put in a leveraged performance much more
impressive than the bullion price itself.
There will be consolidations and corrections along the way, but I
think there will be many gold explorers and developers that rack up
share price gains in the hundreds of percent. That doesn’t mean buying
blindly and never trading. We still need to adjust when a stock gets
overweight and manage risk around major exploration campaigns. The last
few weeks has been a lot more fun in the resource space. I don’t think
the fun’s over yet. Enjoy the ride.
Like any good contrarian, a 10-year bull market makes me alert of
signs of potential trouble. As noted at the start of this editorial, I’m
expecting continues floods of liquidity. That may simply overwhelm
everything else for a while and allow Wall St to keep rallying, come
what may.
That said, a couple of data points recently got my attention. One is
more of a sentiment indicator, seen in the chart below. More than one
wag has joked that the Fed need only worry about Wall St, since the
stock market is the economy now. Turns out there is more than a bit of
truth to that.
The chart shows the US Leading Indicator reading with the level of
the stock market (which is a component of the official Leading
Indicator) removed. As you can see, without Wall St, the indicator
implies zero growth going forward. I’m mainly showing it as evidence of
just how surreal things have become.
The chart above is something to keep an eye on going forward. It
shows weekly State unemployment claims for several major sectors of the
economy. What’s interesting about this chart is that claims have been
climbing rapidly over the past few weeks. Doubly interesting is that the
increase in claims is broad, both within and across several sectors of
the economy.
I take the monthly Non-Farm Payroll number less seriously than most,
because it’s a backward-looking indicator. This move in unemployment
claims looks increasingly like a trend though. It’s now at its highest
level since the Financial Crisis.
It’s not in the danger zone-yet. But its climbing fast. We may need
to start paying more attention to those payroll numbers. If the chart
below isn’t a statistical fluke, we may start seeing negative surprises
in the NFP soon. That won’t hurt the gold price either.
Source and Thanks: https://www.hraadvisory.com/golds-big-picture
Tags: #VRIC Posted in Affinity Metals, All Recent Posts | Comments Off on Gold’s Big Picture SPONSOR: Affinity Metals $AAF.ca $SII.ca $TUD.ca $GTT.ca $AMK.ca $OSK.ca $RKR.ca
Posted by AGORACOM
at 10:37 AM on Thursday, January 16th, 2020
Building codes are a labyrinth of national, state, and municipal
rules. While California since 2015 has required new homes to have the
necessary conduit and service-panel capacity for EV-charging, guidelines
in the rest of the country are spotty. That could soon be fixed because
the International Code Council (ICC) – which provides widely adopted
best practices and standards for construction – approved putting
EV-readiness in its latest guidelines.
The new guidelines equate to a ready-made, consistent national
approach for EV-charging capabilities for new homes and apartment
buildings.
While all states follow the principles outlined by the ICC’s building
codes, the provisions are voluntary until incorporated into state or
local laws. Quartz reports that about half of US states are expected to adopt the ICC’s new EV-readiness requirements.
Forward-looking municipalities – notably Atlanta, Denver, Palo Alto,
and Seattle – already have EV-friendly construction codes in place.
Estimates for the cost of compliance for a newly constructed home vary widely from less than $100 to nearly $1,000.
A 2016 study
pegged the price in San Francisco to be $920 (for a building with 10
parking spaces). But that’s significantly less than adding charging
capabilities after the fact. The same research indicates that
retrofitting sites by expanding electrical panels and adding wiring,
could cost as much as $3,550.
The ICC explains, “The proposed code [now adopted] will allow current
and future EV-owners to avoid the cost of electrical equipment
upgrades, demolition, and permitting for future retrofits.â€
ELECTRIC VEHICLE SUPPLY EQUIPMENT (EVSE). The conductors, including
the ungrounded, grounded, and equipment grounding conductors, and the
Electric Vehicle connectors, attachment plugs, and all other fittings,
devices, power outlets, or apparatus installed specifically for the
purpose of transferring energy between the premises wiring and the
Electric Vehicle.
EV CAPABLE SPACE. Electrical panel capacity and space to support a
minimum 40-ampere, 208/240-volt branch circuit for each EV parking
space, and the installation of raceways, both underground and surface
mounted, to support the EVSE.
EV READY SPACE. A designated parking space which is provided with
one 40-ampere, 208/240-volt dedicated branch circuit for EVSE servicing
Electric Vehicles. The circuit shall terminate in a suitable termination
point such as a receptacle, junction box, or an EVSE, and be located in
close proximity to the proposed location of the EV parking spaces.
While builders will make sure that there’s access to a 240-volt
supply, it’s up to owners or tenants to buy and install the charging
equipment.
The ICC says there will need to be 9.6 million new EV charging ports by 2030, with nearly 80% located in single and multi-family residential buildings. As any EV driver knows, home is where the vast majority of electric-car charging takes place.
Posted by AGORACOM
at 9:18 AM on Wednesday, January 15th, 2020
Vancouver, British Columbia–(Newsfile Corp. – January 15, 2020) – Lomiko Metals Inc.
(TSXV: LMR) would like to cordially invite you to visit us at Booth
#1030 at the Vancouver Resource Investment Conference (VRIC) to be held
at the Vancouver Convention Centre West (1055 Canada Place, Vancouver)
on Sunday January 19 – Monday January 20, 2020.
The Vancouver Resource Investment Conference has been the bellwether
of the junior mining market for the last twenty-five years. It is the
number one source of information for investment trends and ideas,
covering all aspects of the natural resource industry.
Each year, the VRIC hosts over 60 keynote speakers, 350 exhibiting companies and 9000 investors.
Investment thought leaders and wealth influencers provide our
audiences with valuable insights. C-suite company executives covering
every corner of the mineral exploration sector as well as metals, oil
& gas, renewable energy, media and financial services companies are
available to speak one on one. This is a must-attend for investors and
stakeholders in the global mining industry.